INDEX

Explanations

incidents involving violence or harm towards individuals

oai_token-act-pair · gpt-3.5-turbo

phrases related to events involving injury or death

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

ennial

-0.86

CHR

-0.85

}}}

-0.77

issance

-0.76

facebook

-0.74

Row

-0.74

iddles

-0.73

ãĥ¼ãĥ«

-0.73

hi

-0.73

Color

-0.72

POSITIVE LOGITS

 gunfire

1.06

 gunshots

0.97

 ingest

0.96

 gunshot

0.93

 paramedics

0.91

 ingestion

0.90

 untreated

0.90

 sustaining

0.87

 deton

0.87

 downed

0.86

Activations Density 0.227%

incidents involving violence or harm towards individuals

phrases related to events involving injury or death

No Comments

No Known Activations