INDEX

Explanations

statements indicating lack of knowledge or awareness

oai_token-act-pair · gpt-3.5-turbo

statements of lack of knowledge or awareness about specific facts or discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 beware

-0.69

nels

-0.68

nel

-0.65

hani

-0.65

ht

-0.64

pping

-0.64

tick

-0.64

shift

-0.63

ayers

-0.63

claw

-0.63

POSITIVE LOGITS

 satisfactory

1.08

 credible

1.01

 conclusive

0.96

 reliable

0.95

 definitive

0.95

 evidence

0.92

 documented

0.88

 indication

0.87

 instances

0.87

 mention

0.86

Activations Density 0.307%

statements indicating lack of knowledge or awareness

statements of lack of knowledge or awareness about specific facts or discussions

No Comments

No Known Activations