INDEX

Explanations

No clear pattern or meaning found in the activations for Neuron 4

oai_token-act-pair · gpt-3.5-turbo

numerical values or codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

itory

-0.75

 Beir

-0.72

ipation

-0.70

itionally

-0.66

ãĥ¤

-0.66

 EntityItem

-0.65

arians

-0.65

angible

-0.65

imaru

-0.64

icas

-0.64

POSITIVE LOGITS

teenth

1.31

teen

1.22

th

0.95

0.92

0.91

0.88

0.87

0.86

Activations Density 0.064%

No clear pattern or meaning found in the activations for Neuron 4

numerical values or codes

No Comments

No Known Activations