INDEX

Explanations

prepositions followed by specific keywords or phrases

oai_token-act-pair · gpt-3.5-turbo

phrases indicating specific positioning or timing

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ories

-0.83

iple

-0.78

/-

-0.71

vari

-0.68

crew

-0.66

mt

-0.66

rams

-0.66

RAM

-0.66

ply

-0.66

intend

-0.66

POSITIVE LOGITS

cue

0.77

 doorstep

0.65

 Springer

0.65

 Shooter

0.61

 noses

0.60

 Grimes

0.60

 Gomez

0.59

 borders

0.56

ito

0.56

 Border

0.55

Activations Density 0.085%

prepositions followed by specific keywords or phrases

phrases indicating specific positioning or timing

No Comments

No Known Activations