INDEX

Explanations

phrases related to citations or references

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

Data Type

float32

Hook Name

blocks.12.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

帖最后由

-0.84

NUMX

-0.81

новниш

-0.78

expandindo

-0.77

 Phaser

-0.77

 numberWith

-0.73

nerability

-0.72

 שוליים

-0.72

 NgModule

-0.71

\{\\

-0.70

POSITIVE LOGITS

see

0.58

 reprend

0.57

 Consultez

0.55

Mimi

0.55

 Confira

0.54

Suggest

0.53

 Consulte

0.53

 hjælp

0.52

the

0.51

 råd

0.51

Activations Density 0.002%

phrases related to citations or references

No Comments

No Known Activations