INDEX

Explanations

conditional statements starting with "if" or indicating a contrast with "albeit"

oai_token-act-pair · gpt-3.5-turbo

conditional or contrasting phrases

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ESE

-0.87

 GOODMAN

-0.83

onds

-0.75

enter

-0.74

rients

-0.73

sbm

-0.70

idth

-0.70

ascus

-0.69

thia

-0.69

vantage

-0.68

POSITIVE LOGITS

 imperfect

1.24

 fleeting

1.23

 incomplete

1.22

 modest

1.21

 somewhat

1.19

 reluctantly

1.18

 slightly

1.15

 unintentional

1.13

 tentative

1.13

 anecdotal

1.11

Activations Density 0.163%

conditional statements starting with "if" or indicating a contrast with "albeit"

conditional or contrasting phrases

No Comments

No Known Activations