INDEX

Explanations

symbols and Twitter usernames

oai_token-act-pair · gpt-3.5-turbo

references to specific people or entities, particularly in a formal or notable context

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

pee

-0.85

EE

-0.81

-0.75

zsche

-0.75

 Erie

-0.74

cer

-0.74

tery

-0.74

 upgr

-0.74

pees

-0.73

 Wichita

-0.73

POSITIVE LOGITS

 Martin

1.92

Martin

1.79

 Mart

1.24

mart

1.13

Mel

1.10

Mart

1.08

 mart

1.06

 Lambert

1.03

 Marino

0.99

 Malt

0.98

Activations Density 0.358%

symbols and Twitter usernames

references to specific people or entities, particularly in a formal or notable context

No Comments

No Known Activations