INDEX

Explanations

references to minimum quantities or values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.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

 only

-0.42

 humanidade

-0.42

istoitu

-0.40

 actualité

-0.40

JNIEnv

-0.40

 bezeichneter

-0.38

gjenge

-0.38

 infatti

-0.38

 Various

-0.37

 verschied

-0.37

POSITIVE LOGITS

 partially

0.82

 partly

0.70

 Partially

0.68

 EconPapers

0.66

 parcialmente

0.65

 moderately

0.61

 teilweise

0.60

partly

0.60

 partial

0.59

 rudi

0.59

Activations Density 0.193%

references to minimum quantities or values

No Comments

No Known Activations