INDEX

Explanations

It appears that neuron 4 does not activate for any tokens in the provided dataset, suggesting that the neuron is either not functioning correctly or that its specific search criteria were not present in the text

oai_token-act-pair · gpt-4-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ciating

-0.79

idan

-0.74

zai

-0.73

-0.72

 subscrib

-0.69

":[

-0.68

Laughs

-0.68

ItemThumbnailImage

-0.67

inav

-0.67

Ĭ±

-0.67

POSITIVE LOGITS

NEO

0.74

 CONT

0.69

 Tacoma

0.68

Moz

0.67

CHO

0.65

 Cherokee

0.63

FTC

0.62

TO

0.61

fort

0.60

 âī

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.