INDEX

Explanations

words related to persistent or lasting situations or conditions

oai_token-act-pair · gpt-3.5-turbo

terms related to prolonged presence or effects

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ħĭ

-0.77

raid

-0.76

reme

-0.75

icient

-0.75

portion

-0.73

mits

-0.72

ducers

-0.72

abet

-0.71

fit

-0.70

rams

-0.70

POSITIVE LOGITS

 lingering

1.15

 remnants

0.87

 linger

0.86

 ling

0.84

 Jagu

0.77

 adolesc

0.77

aft

0.75

INGTON

0.75

 suspense

0.74

Ire

0.71

Activations Density 0.008%

words related to persistent or lasting situations or conditions

terms related to prolonged presence or effects

No Comments

No Known Activations