INDEX

Explanations

instances of the word "once" followed by actions or conditions involving "you."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

StructEnd

-0.55

舺

-0.52

Попис

-0.51

RTLU

-0.49

MockMvc

-0.47

IsMutable

-0.44

parsedMessage

-0.44

adox

-0.42

IBOutlet

-0.42

RIOS

-0.42

POSITIVE LOGITS

 Sobald

0.64

 Once

0.60

Once

0.59

 sobald

0.52

一旦

0.48

 readily

0.46

 once

0.43

 Setelah

0.42

 attained

0.41

Upon

0.41

Activations Density 0.153%

instances of the word "once" followed by actions or conditions involving "you."

No Comments

No Known Activations