INDEX

Explanations

phrases related to ongoing actions or processes

oai_token-act-pair · gpt-3.5-turbo

terms related to ongoing processes or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Ging

-0.68

Pix

-0.67

cause

-0.67

express

-0.63

Psy

-0.63

pop

-0.62

 Coup

-0.61

ults

-0.60

ins

-0.60

don

-0.59

POSITIVE LOGITS

iever

0.72

Ø©

0.70

ESE

0.70

ALSE

0.69

 reversible

0.68

 dock

0.66

////////

0.65

 limbo

0.65

earable

0.64

Øª

0.64

Activations Density 0.112%

phrases related to ongoing actions or processes

terms related to ongoing processes or activities

No Comments

No Known Activations