INDEX

Explanations

ed adjectives

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

sii

-0.97

 laun

-0.96

 ?...

-0.95

tnt

-0.95

 lara

-0.94

prs

-0.94

 secon

-0.93

 wald

-0.91

 zimmer

-0.89

 bose

-0.89

POSITIVE LOGITS

 ***!

0.61

 fuck

0.59

fuck

0.57

<bos>

0.55

ajemen

0.55

 проєкту

0.52

 continuant

0.51

 ducato

0.51

Fuck

0.49

Activations Density 0.078%

ed adjectives

No Comments

No Known Activations