INDEX

Explanations

instances of text related to investigations being conducted

oai_token-act-pair · gpt-3.5-turbo

references to investigations or inquiries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ãĤª

-0.76

ãĥĪ

-0.75

icio

-0.72

Ã±a

-0.69

translation

-0.69

Zone

-0.65

layout

-0.65

anguage

-0.64

ritical

-0.64

ranged

-0.64

POSITIVE LOGITS

 whether

1.37

 allegations

1.32

why

1.10

 claims

1.01

 complaints

1.00

how

0.95

 accusations

0.95

whether

0.95

 incidents

0.94

 irregularities

0.93

Activations Density 0.096%

instances of text related to investigations being conducted

references to investigations or inquiries

No Comments

No Known Activations