INDEX

Explanations

words related to ratings and reviews

oai_token-act-pair · gpt-3.5-turbo

numerical ratings and evaluations of products

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

bleacher

-0.74

 Chomsky

-0.74

 tacit

-0.72

enhagen

-0.71

Åį

-0.70

 Consortium

-0.70

amph

-0.70

 hypothetical

-0.69

—"

-0.68

Å«

-0.67

POSITIVE LOGITS

 LOVE

1.04

Rated

1.02

review

0.95

LO

0.94

 Review

0.90

BuyableInstoreAndOnline

0.88

 reviewers

0.88

Review

0.87

 reviewer

0.86

 alot

0.86

Activations Density 0.635%

words related to ratings and reviews

numerical ratings and evaluations of products

No Comments

No Known Activations