INDEX

Explanations

reviews and ratings of products or services

oai_token-act-pair · gpt-3.5-turbo

product ratings and evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 satir

-1.00

ulkan

-0.75

 enshr

-0.70

govtrack

-0.69

 Jinn

-0.69

 incentiv

-0.69

 Wikimedia

-0.68

 Confederation

-0.64

 broadcasting

-0.64

href

-0.63

POSITIVE LOGITS

Rated

1.34

 Bought

1.24

 Purch

1.12

Excellent

1.05

 Received

1.04

 Review

1.03

 Customer

1.02

 Excellent

1.02

 Highly

1.01

Fit

1.01

Activations Density 0.361%

reviews and ratings of products or services

product ratings and evaluations

No Comments

No Known Activations