INDEX

Explanations

references to legal proceedings or police investigations

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

mī

-0.73

 izvē

-0.57

 philosophic

-0.56

UnusedPrivate

-0.54

vī

-0.53

 Marquette

-0.52

 vairāk

-0.52

 ferric

-0.52

 Doran

-0.51

 liberality

-0.48

POSITIVE LOGITS

 Watson

1.27

Watson

1.13

 Holmes

0.98

 WATSON

0.93

 Sherlock

0.89

Holmes

0.88

Sherlock

0.79

IBM

0.72

IBM

0.68

 Conan

0.64

Activations Density 0.265%

references to legal proceedings or police investigations

No Comments

No Known Activations