INDEX

Explanations

phrases indicating a need or requirement

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

 Peres

-0.76

 betrek

-0.75

 vlas

-0.72

 hôtes

-0.72

 auroit

-0.70

 colgantes

-0.69

 Numerade

-0.69

hdashline

-0.69

ölf

-0.69

PRWEB

-0.69

POSITIVE LOGITS

 needs

0.96

be

0.90

 need

0.90

 must

0.90

Neces

0.88

 gotta

0.82

zzleHttp

0.78

 Neces

0.77

 urgently

0.77

 måste

0.76

Activations Density 0.042%

phrases indicating a need or requirement

No Comments

No Known Activations