INDEX

Explanations

references to locations and political events or movements

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

 alkoh

-0.73

 maksi

-0.65

pü

-0.63

 kompati

-0.62

 mikrofon

-0.61

первых

-0.60

 kompres

-0.59

 kristal

-0.58

 kosme

-0.58

 ujednoznacz

-0.57

POSITIVE LOGITS

 SIMILAR

0.82

 similar

0.79

 teresa

0.75

ftu

0.75

similar

0.74

 jorge

0.74

 fign

0.74

 paula

0.74

<^

0.71

 gabriel

0.70

Activations Density 0.086%

references to locations and political events or movements

No Comments

No Known Activations