INDEX

Explanations

compound words created by joining two smaller words together

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to contrasts or oppositions, particularly in social contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WER

-0.78

master

-0.75

 Maid

-0.75

 Metropolitan

-0.71

erness

-0.69

 Annotations

-0.68

REDACTED

-0.65

 Reserve

-0.64

DEN

-0.64

bucks

-0.63

POSITIVE LOGITS

anging

1.16

isions

1.13

ining

1.12

asive

1.08

icably

1.05

osed

1.02

osing

1.01

odied

1.00

aced

0.99

isting

0.98

Activations Density 0.029%

compound words created by joining two smaller words together

words and phrases related to contrasts or oppositions, particularly in social contexts

No Comments

No Known Activations