INDEX

Explanations

codes or identifiers in a structured format

oai_token-act-pair · gpt-3.5-turbo

numerical identifiers or codes

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

quartered

-0.79

drawn

-0.74

accessible

-0.73

arching

-0.73

swick

-0.72

ufact

-0.71

estine

-0.70

oggles

-0.69

illard

-0.68

amination

-0.67

POSITIVE LOGITS

0.96

0.86

chell

0.85

ĸļ

0.80

=#

0.79

qqa

0.77

ETH

0.76

00200000

0.76

0.75

Activations Density 0.020%

codes or identifiers in a structured format

numerical identifiers or codes

No Comments

No Known Activations