INDEX

Explanations

coding structures related to class and method definitions, typically in programming environments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

[]:

-0.60

setopt

-0.51

 doInBackground

-0.50

hearsed

-0.49

帖最后由

-0.47

Referencies

-0.47

indexPath

-0.46

APH

-0.45

querySelector

-0.45

Референце

-0.45

POSITIVE LOGITS

<eos>

0.99

Portály

0.69

UserScript

0.66

MemoryWarning

0.64

=$?

0.63

expandindo

0.63

Brainz

0.63

+:+

0.61

udsman

0.61

neurial

0.57

Activations Density 0.002%

coding structures related to class and method definitions, typically in programming environments

No Comments

No Known Activations