INDEX

Explanations

7

np_max-act-logits · gemini-2.5-flash-lite

I cannot determine what this neuron is looking for based on the provided data, as it shows zero activation values across all tokens in every example.

oai_token-act-pair · claude-4-5-haiku Triggered by @enricobottazzi

the numeral 7, especially within phone numbers or numerical strings.

oai_token-act-pair · gpt-5 Triggered by @enricobottazzi

New Auto-Interp

Configuration

google/gemma-scope-2-27b-it/transcoder_all/layer_57_width_262k_l0_small_affine

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ALITY

0.40

APD

0.37

 temu

0.35

(°

0.34

 برگ

0.33

cké

0.33

té

0.33

]++;

0.33

TES

0.33

ства

0.32

POSITIVE LOGITS

faf

0.41

luoromethyl

0.39

ța

0.38

Ϯ

0.37

haw

0.36

lut

0.36

্রো

0.36

rata

0.35

delicious

0.35

∣

0.35

Activations Density 0.002%

7

I cannot determine what this neuron is looking for based on the provided data, as it shows zero activation values across all tokens in every example.

the numeral 7, especially within phone numbers or numerical strings.

No Comments

No Known Activations

7

I cannot determine what this neuron is looking for based on the provided data, as it shows zero activation values across all tokens in every example.

the numeral 7, especially within phone numbers or numerical strings.

No Comments

No Known Activations