INDEX

Explanations

the word "that" and references to individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 itſelf

-1.96

 myſelf

-1.89

Efq

-1.77

 purpoſe

-1.74

 ―――――

-1.74

 Majefty

-1.69

 Monfieur

-1.68

 doubtnut

-1.68

 himſelf

-1.63

 themſelves

-1.61

POSITIVE LOGITS

1.07

 That

1.07

That

1.06

<eos>

0.97

0.95

 что

0.93

0.91

 that

0.85

Activations Density 0.118%

the word "that" and references to individuals

No Comments

No Known Activations