INDEX

Explanations

words related to unbreakable or unpenetrable situations or barriers

oai_token-act-pair · gpt-3.5-turbo

prefixes and suffixes in words related to barriers, boundaries, or limitations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 initials

-0.69

minster

-0.68

 Patton

-0.63

 goodbye

-0.61

 Bans

-0.61

 boarding

-0.59

fork

-0.58

spin

-0.58

 reckoning

-0.57

works

-0.56

POSITIVE LOGITS

iture

1.05

acy

1.05

ibly

1.04

iencies

1.03

icative

0.99

icate

0.96

ications

0.94

itures

0.94

ication

0.93

ously

0.93

Activations Density 0.195%

words related to unbreakable or unpenetrable situations or barriers

prefixes and suffixes in words related to barriers, boundaries, or limitations

No Comments

No Known Activations