INDEX

Explanations

words related to regulations and being regulated

oai_token-act-pair · gpt-3.5-turbo

terms related to regulatory frameworks and regulations

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

hur

-0.87

lay

-0.83

issues

-0.81

odor

-0.80

ARC

-0.77

alter

-0.77

now

-0.75

clerosis

-0.74

 Takeru

-0.73

PU

-0.73

POSITIVE LOGITS

aily

0.86

 subur

0.84

ategory

0.80

 tradem

0.79

 millenn

0.78

uled

0.78

 shorth

0.78

 overseen

0.77

 territ

0.76

 adolesc

0.76

Activations Density 0.047%

words related to regulations and being regulated

terms related to regulatory frameworks and regulations

No Comments

No Known Activations