INDEX

Explanations

mentions related to accusations or legal issues

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 fédé

-0.67

 alliés

-0.58

 délib

-0.55

 semblables

-0.53

 présidenti

-0.51

 traités

-0.50

 inégal

-0.50

Etimo

-0.49

 destinées

-0.48

 récents

-0.48

POSITIVE LOGITS

who

0.70

those

0.70

 those

0.65

Those

0.63

 whom

0.63

who

0.60

 THOSE

0.58

 mène

0.58

 tanong

0.58

 Those

0.57

Activations Density 0.088%

mentions related to accusations or legal issues

No Comments

No Known Activations