INDEX

Explanations

numerical values followed by a specific word or phrase

oai_token-act-pair · gpt-3.5-turbo

numeric values, particularly those related to statistics or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 tradem

-1.04

 millenn

-1.01

¥ŀ

-0.90

umn

-0.87

 satell

-0.78

ĻĤ

-0.77

holders

-0.77

psey

-0.76

ntil

-0.76

secut

-0.76

POSITIVE LOGITS

0.99

rd

0.91

 ILCS

0.88

0.86

0.83

çļ

0.83

Prince

0.82

Activations Density 0.050%

numerical values followed by a specific word or phrase

numeric values, particularly those related to statistics or identifiers

No Comments

No Known Activations