INDEX

Explanations

sequences of numbers and symbols likely related to coding or technical data

oai_token-act-pair · gpt-3.5-turbo

numerical patterns and sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

tions

-0.74

ername

-0.70

@#&

-0.69

oleon

-0.66

================================================================

-0.66

 Forth

-0.65

theless

-0.65

 Username

-0.65

 Mental

-0.64

nostic

-0.64

POSITIVE LOGITS

crop

0.79

icons

0.75

chrom

0.75

azo

0.72

)=(

0.71

usercontent

0.69

cdn

0.65

original

0.65

bleacher

0.64

"><

0.64

Activations Density 0.049%

sequences of numbers and symbols likely related to coding or technical data

numerical patterns and sequences

No Comments

No Known Activations