INDEX

Explanations

proper nouns related to political figures and entities

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals or entities with the suffix "ano."

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

nesday

-0.90

owship

-0.80

tered

-0.75

 sockets

-0.74

monton

-0.70

drawn

-0.70

brain

-0.69

rences

-0.69

respective

-0.69

swick

-0.66

POSITIVE LOGITS

jo

1.17

zzi

1.06

ahu

0.96

qua

0.94

ose

0.92

zzo

0.90

zo

0.90

hazard

0.88

ichi

0.88

heim

0.83

Activations Density 0.021%

proper nouns related to political figures and entities

references to specific individuals or entities with the suffix "ano."

No Comments

No Known Activations