INDEX

Explanations

sentences with vocabulary related to instruction and explanation

oai_token-act-pair · gemini-2.0-flash

review

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enderror

-0.63

 Audiodateien

-0.55

 صوتيه

-0.54

estrado

-0.53

fetera

-0.53

 XCTest

-0.53

})}\

-0.52

)':

-0.50

LayoutPanel

-0.50

 خارجية

-0.50

POSITIVE LOGITS

 review

1.87

review

1.68

 Review

1.68

 REVIEW

1.60

Review

1.59

 reviewed

1.44

REVIEW

1.43

 reviewing

1.41

 reviews

1.36

reviewed

1.23

Activations Density 3.313%

sentences with vocabulary related to instruction and explanation

review

No Comments

No Known Activations