INDEX

Explanations

words related to being isolated or cut off from others

oai_token-act-pair · gpt-3.5-turbo

instances of the prefix "ins," often indicating a condition or quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

DragonMagazine

-0.73

ARK

-0.73

ries

-0.70

XD

-0.68

ãĥ£

-0.66

 Master

-0.64

 Kavanaugh

-0.63

 Rumble

-0.62

ttes

-0.61

 Stafford

-0.61

POSITIVE LOGITS

ensible

1.21

urances

1.09

istent

1.08

ufficient

1.03

pite

1.03

ensibly

1.01

criptions

0.99

semb

0.97

uper

0.96

iders

0.95

Activations Density 0.015%

words related to being isolated or cut off from others

instances of the prefix "ins," often indicating a condition or quality

No Comments

No Known Activations