INDEX

Explanations

keywords related to programming code or specific identifiers

oai_token-act-pair · gpt-3.5-turbo

structured data or lists related to specific topics

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

awa

-0.81

hement

-0.80

 fors

-0.74

 appe

-0.70

 application

-0.70

 tired

-0.69

 favoured

-0.69

 agre

-0.66

 tink

-0.65

romeda

-0.65

POSITIVE LOGITS

Rail

0.90

Legendary

0.84

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

0.81

Population

0.81

Dist

0.80

Staff

0.79

Joined

0.78

================================================================

0.77

croft

0.77

Age

0.76

Activations Density 0.300%

keywords related to programming code or specific identifiers

structured data or lists related to specific topics

No Comments

No Known Activations