INDEX

Explanations

technical or procedural words and phrases

oai_token-act-pair · gpt-3.5-turbo

phrases related to warnings, particularly those related to safety or health

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 sqor

-0.81

idays

-0.79

leground

-0.76

osures

-0.74

assadors

-0.73

rities

-0.73

rosso

-0.66

reddits

-0.66

ournaments

-0.65

enaries

-0.64

POSITIVE LOGITS

 chlorine

0.69

 deton

0.64

 Symbol

0.64

 corro

0.64

 sorcery

0.63

 photon

0.63

 symbol

0.63

 engraved

0.60

 hardness

0.58

 Morse

0.57

Activations Density 1.510%

technical or procedural words and phrases

phrases related to warnings, particularly those related to safety or health

No Comments

No Known Activations