INDEX

Explanations

nutrients

oai_token-act-pair · gpt-3.5-turbo

terms related to nutrition and nutrients

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cker

-0.74

hair

-0.74

prototype

-0.72

ker

-0.71

door

-0.70

cks

-0.68

vu

-0.66

 hands

-0.66

bley

-0.65

king

-0.64

POSITIVE LOGITS

ritional

1.53

rients

1.44

 nutrients

1.17

 nutrit

1.08

 nutrient

1.03

rient

1.01

 carbohyd

0.97

rition

0.96

orie

0.96

 nutritional

0.93

Activations Density 0.014%

nutrients

terms related to nutrition and nutrients

No Comments

No Known Activations