INDEX

Explanations

technical terms related to protection or security

oai_token-act-pair · gpt-3.5-turbo

terms related to protection and safety

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 discharge

-0.69

 arsen

-0.61

 accrued

-0.61

olyn

-0.58

 disclaim

-0.57

deal

-0.57

 prehistoric

-0.56

MIA

-0.55

 waning

-0.55

 onset

-0.54

POSITIVE LOGITS

Pub

1.48

Prot

1.44

DAQ

0.89

Els

0.86

Priv

0.81

Nap

0.80

Translation

0.76

URI

0.76

Flags

0.75

%"

0.73

Activations Density 0.007%

technical terms related to protection or security

terms related to protection and safety

No Comments

No Known Activations