INDEX

Explanations

references to periods of time in the past

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the time frame of events, specifically referencing the past

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Franch

-0.74

hyde

-0.71

NEY

-0.71

iform

-0.66

amber

-0.66

UGH

-0.65

lee

-0.65

wagon

-0.64

ithmetic

-0.61

witz

-0.61

POSITIVE LOGITS

ebin

1.38

decade

1.19

 week

0.94

 millennium

0.93

 century

0.89

 month

0.88

orate

0.88

 tense

0.88

 semester

0.83

few

0.82

Activations Density 0.023%

references to periods of time in the past

phrases indicating the time frame of events, specifically referencing the past

No Comments

No Known Activations