INDEX

Explanations

Given the variety of activations, it seems like this neuron is difficult to summarize in a single sentence

oai_token-act-pair · gpt-3.5-turbo

specific proper nouns or names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Hein

-0.84

 Bauer

-0.80

Hug

-0.80

iel

-0.77

hered

-0.75

 enthusi

-0.74

 Bees

-0.71

 Goose

-0.70

gra

-0.70

iveness

-0.70

POSITIVE LOGITS

 Tate

0.89

tmp

0.89

stant

0.88

wart

0.87

ton

0.87

DT

0.85

qt

0.83

ot

0.83

tty

0.81

dt

0.81

Activations Density 0.252%

Given the variety of activations, it seems like this neuron is difficult to summarize in a single sentence

specific proper nouns or names

No Comments

No Known Activations