INDEX

Explanations

words related to incidents involving unarmed individuals, particularly in cases involving law enforcement

oai_token-act-pair · gpt-3.5-turbo

references to unarmed individuals, particularly in contexts involving violence or police interactions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

nosis

-0.78

emetery

-0.77

MER

-0.76

_-

-0.73

Magazine

-0.72

DCS

-0.72

omics

-0.71

Reward

-0.67

RAM

-0.67

URI

-0.67

POSITIVE LOGITS

 unarmed

1.07

 bystand

1.02

 uprising

0.83

 disarm

0.82

 bystanders

0.80

 insurrection

0.77

 civilians

0.76

 manslaughter

0.76

 strikers

0.73

ives

0.73

Activations Density 0.018%

words related to incidents involving unarmed individuals, particularly in cases involving law enforcement

references to unarmed individuals, particularly in contexts involving violence or police interactions

No Known Activations