INDEX

Explanations

words relating to classifications, statuses, and modifications in various contexts

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

 ſche

-0.87

:✨

-0.87

WhereInput

-0.84

Personensuche

-0.83

 iſt

-0.81

ſelf

-0.79

 Jefus

-0.78

ſelves

-0.78

 beſ

-0.77

 raiſ

-0.77

POSITIVE LOGITS

0.67

and

0.59

0.58

менте

0.57

UVWXYZ

0.55

 formen

0.55

batore

0.54

0.52

postdata

0.52

mante

0.52

Activations Density 1.263%

words relating to classifications, statuses, and modifications in various contexts

No Comments

No Known Activations