INDEX

Explanations

mentions of speakers or speaking engagements

oai_token-act-pair · gpt-3.5-turbo

references to "Speakers" or "Speaker" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

etheless

-0.86

 guiActiveUnfocused

-0.80

sym

-0.74

 Leilan

-0.70

MpServer

-0.69

ORED

-0.69

enegger

-0.68

imaru

-0.66

 Pradesh

-0.63

-0.61

POSITIVE LOGITS

eches

1.06

lling

1.06

edy

0.97

ciation

0.95

eding

0.95

akers

0.92

ller

0.92

cker

0.91

culation

0.89

llers

0.88

Activations Density 0.016%

mentions of speakers or speaking engagements

references to "Speakers" or "Speaker" in various contexts

No Comments

No Known Activations