INDEX

Explanations

individuals or professions followed by a numeric rating or description

oai_token-act-pair · gpt-3.5-turbo

roles or titles associated with individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

EStreamFrame

-0.75

ween

-0.74

ÙĲ

-0.71

Roaming

-0.66

selves

-0.64

fw

-0.64

Translation

-0.63

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.63

izoph

-0.62

FREE

-0.62

POSITIVE LOGITS

 Geoff

1.04

 Shaun

1.04

 warns

1.03

 extraord

1.03

 defends

1.02

 Jerry

1.01

 Patrick

1.00

 Gina

1.00

 Tony

1.00

 Stephen

1.00

Activations Density 0.267%

individuals or professions followed by a numeric rating or description

roles or titles associated with individuals

No Comments

No Known Activations