INDEX

Explanations

phrases related to email or messaging

oai_token-act-pair · gpt-3.5-turbo

words related to emergency medical services or emergency situations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Antarctic

-0.66

Tat

-0.66

 Indigenous

-0.64

 Chatt

-0.62

 Penguin

-0.62

STE

-0.62

tan

-0.62

 Rapp

-0.61

 Sport

-0.59

skill

-0.59

POSITIVE LOGITS

ems

1.30

pace

1.03

heet

1.01

itage

0.93

ayer

0.92

sonian

0.88

rums

0.87

emed

0.86

ighty

0.85

lectic

0.84

Activations Density 0.006%

phrases related to email or messaging

words related to emergency medical services or emergency situations

No Comments

No Known Activations