INDEX

Explanations

words related to identification such as "card," "forms," and "number."

oai_token-act-pair · gpt-3.5-turbo

references to various forms of identification documents and requirements

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

ovie

-1.04

aimon

-0.87

uden

-0.84

^^^^

-0.81

ategory

-0.77

orld

-0.74

scill

-0.72

peak

-0.72

imeo

-0.72

~~~~~~~~

-0.70

POSITIVE LOGITS

 identification

1.10

 credentials

1.07

 badges

1.03

IDs

1.01

 booths

0.95

 badge

0.92

 cards

0.92

 scanners

0.91

 papers

0.91

 credential

0.89

Activations Density 0.061%

words related to identification such as "card," "forms," and "number."

references to various forms of identification documents and requirements

No Comments

No Known Activations