INDEX

Explanations

phrases related to studying or analysis

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

 maneu

-1.66

 affor

-1.60

 emphat

-1.59

 disagre

-1.58

 increa

-1.55

 shenan

-1.52

 intersper

-1.52

 reluct

-1.49

 depic

-1.44

fuf

-1.42

POSITIVE LOGITS

 Instead

1.10

Instead

1.00

 instead

0.93

 Therefore

0.82

instead

0.78

tdessen

0.76

Therefore

0.72

 Rather

0.72

 prefer

0.71

SafeAreaView

0.68

Activations Density 0.504%

phrases related to studying or analysis

No Comments

No Known Activations