INDEX

Explanations

words related to stealth or hidden presence

oai_token-act-pair · gpt-3.5-turbo

words related to lurking and crouching actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ocr

-0.69

 biased

-0.68

 Luther

-0.68

 Samar

-0.65

ripp

-0.63

emale

-0.61

ractor

-0.60

regn

-0.58

ract

-0.56

 bias

-0.56

POSITIVE LOGITS

ched

1.19

ches

1.09

kish

1.07

ching

1.05

chers

1.04

cks

1.02

ks

1.00

chery

1.00

ked

0.96

kers

0.95

Activations Density 0.078%

words related to stealth or hidden presence

words related to lurking and crouching actions

No Comments

No Known Activations