INDEX

Explanations

instances of speech and expression, including forms of the word "speak."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 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

 Etapa

-0.79

[--

-0.79

 Noruega

-0.74

 Jefus

-0.73

satisfied

-0.73

 ſche

-0.72

']))

-0.72

waltung

-0.69

 Norvège

-0.69

 Avez

-0.68

POSITIVE LOGITS

 speak

1.90

 Speak

1.82

 speaking

1.73

 speaks

1.69

Speak

1.68

 spoke

1.64

 spoken

1.54

 speaker

1.54

 speakers

1.51

speak

1.48

Activations Density 0.063%

instances of speech and expression, including forms of the word "speak."

No Comments

No Known Activations