INDEX

Explanations

mention of endings or conclusions

oai_token-act-pair · gpt-3.5-turbo

phrases related to finality or conclusions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

BLIC

-0.80

onson

-0.69

ickr

-0.66

ub

-0.63

ufact

-0.57

 suscept

-0.57

 Awesome

-0.56

æ©Ł

-0.56

Interested

-0.56

 peers

-0.56

POSITIVE LOGITS

owment

1.32

angering

1.11

orph

0.96

game

0.96

ogeneity

0.89

angered

0.88

angers

0.86

bringer

0.83

ocrine

0.81

usive

0.80

Activations Density 0.055%

mention of endings or conclusions

phrases related to finality or conclusions

No Comments

No Known Activations