INDEX

Explanations

explicit conversations between individuals

oai_token-act-pair · gpt-3.5-turbo

occurrences of conversations and dialogue

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Tro

-0.75

ress

-0.69

rip

-0.68

Finish

-0.67

WIN

-0.64

nic

-0.63

sole

-0.62

sk

-0.62

ressing

-0.62

RET

-0.62

POSITIVE LOGITS

 conversations

1.29

uggest

1.08

 discussions

1.07

poons

1.06

ynthesis

1.06

hops

1.06

 chats

0.96

hips

0.95

afety

0.93

ometimes

0.92

Activations Density 0.012%

explicit conversations between individuals

occurrences of conversations and dialogue

No Comments

No Known Activations