INDEX

Explanations

pieces of information or objects within a context

oai_token-act-pair · gpt-3.5-turbo

references to individual items or components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Predators

-0.75

elsius

-0.71

rawdownloadcloneembedreportprint

-0.66

ministic

-0.64

acebook

-0.64

sed

-0.61

croft

-0.60

few

-0.59

 Pione

-0.59

 Monitor

-0.59

POSITIVE LOGITS

meal

2.10

work

1.03

umen

0.92

wise

0.90

 piece

0.80

 pieces

0.80

icles

0.80

pin

0.77

 thereof

0.77

ridges

0.76

Activations Density 0.040%

pieces of information or objects within a context

references to individual items or components

No Comments

No Known Activations