INDEX

Explanations

references to comparison and examples in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

rungsseite

-0.74

parsedMessage

-0.60

 raiſ

-0.57

 externi

-0.57

FieldOffsetTable

-0.56

 initComponents

-0.55

<",

-0.54

StoryboardSegue

-0.54

 faſt

-0.53

 itſelf

-0.52

POSITIVE LOGITS

 подоб

0.56

こういう

0.52

 solcher

0.49

 such

0.47

 like

0.46

这类

0.46

 solchen

0.46

这样的

0.45

 solche

0.44

こういった

0.43

Activations Density 0.346%

references to comparison and examples in discussions

No Comments

No Known Activations