INDEX

Explanations

organizations and services names

oai_token-act-pair · gpt-3.5-turbo

mentions of various government services and agencies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uran

-0.66

orne

-0.66

caps

-0.66

yle

-0.66

jad

-0.64

ussen

-0.64

¿½

-0.62

urate

-0.62

ividual

-0.62

paren

-0.62

POSITIVE LOGITS

 Employees

1.11

 Provider

0.94

 Prosecut

0.80

 Desk

0.79

 Agency

0.77

 Provided

0.76

 Officers

0.72

 Corps

0.71

 acad

0.70

 Agents

0.70

Activations Density 0.031%

organizations and services names

mentions of various government services and agencies

No Comments

No Known Activations