INDEX

Explanations

actions or processes related to evaluating or managing tasks and information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

DOWN

-0.47

down

-0.43

Sil

-0.42

Down

-0.41

 cabo

-0.40

 Down

-0.40

 tartalomajánló

-0.39

Sil

-0.39

hinter

-0.38

Mie

-0.38

POSITIVE LOGITS

 things

0.85

webElementXpaths

0.74

 věci

0.66

 dingen

0.63

 stuff

0.60

 THINGS

0.59

SourceChecksum

0.57

 everything

0.56

AxisAlignment

0.56

 Things

0.56

Activations Density 0.273%

actions or processes related to evaluating or managing tasks and information

No Comments

No Known Activations