INDEX

Explanations

It appears that Neuron 4 has no clear activation pattern as all listed activations have a value of 0, indicating that it does not respond to any of the tokens in the given data

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

 Canad

-0.69

 Macy

-0.68

ulous

-0.66

pta

-0.62

utic

-0.62

uto

-0.62

 Yelp

-0.62

 Newsp

-0.62

 Satisf

-0.62

nant

-0.61

POSITIVE LOGITS

quished

0.77

rior

0.72

fram

0.71

Dear

0.71

ridges

0.70

iless

0.68

åº

0.68

zes

0.67

Pros

0.67

Sold

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.