INDEX

Explanations

phrases indicating strong opinions or directives

oai_token-act-pair · gpt-3.5-turbo

sentences that signify conclusions or definitive statements

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

 helicop

-0.81

 skelet

-0.80

 unden

-0.79

forth

-0.70

lihood

-0.69

 enthusi

-0.68

¥ŀ

-0.65

 gobl

-0.65

gren

-0.64

 corrid

-0.64

POSITIVE LOGITS

 Transcript

0.98

 Specifically

0.96

 Their

0.93

 They

0.91

 Whether

0.88

 Related

0.87

 Topics

0.85

 Though

0.83

 Typically

0.83

 Yesterday

0.83

Activations Density 0.366%

phrases indicating strong opinions or directives

sentences that signify conclusions or definitive statements

No Comments

No Known Activations