INDEX

Explanations

mentions of physical damage or harm

oai_token-act-pair · gpt-3.5-turbo

mentions of "damage" and its related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

zsche

-0.74

zee

-0.72

ramid

-0.72

rams

-0.69

atorial

-0.66

liner

-0.66

ellar

-0.66

tre

-0.65

ounty

-0.64

Screen

-0.61

POSITIVE LOGITS

 inflicted

1.08

 damage

1.04

 mitigation

0.92

damage

0.85

 wrought

0.84

 damaged

0.81

 havoc

0.78

 horm

0.77

 limitation

0.74

 wounds

0.73

Activations Density 0.018%

mentions of physical damage or harm

mentions of "damage" and its related contexts

No Comments

No Known Activations