INDEX

Explanations

Haven't been able to detect a clear pattern in the provided activations for neuron 4 - further analysis might be needed

oai_token-act-pair · gpt-3.5-turbo

the word "wouldn't" and its variations, indicating skepticism or hypothetical scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ULT

-0.70

 Proced

-0.66

 Gutenberg

-0.58

PI

-0.57

gaard

-0.57

dimensional

-0.57

Offline

-0.57

 Casting

-0.56

 Learning

-0.56

 Butt

-0.56

POSITIVE LOGITS

't

1.29

geon

0.97

atically

0.85

terness

0.85

geons

0.82

agy

0.82

acies

0.80

¹

0.79

etsk

0.78

ģĸ

0.77

Activations Density 0.017%

Haven't been able to detect a clear pattern in the provided activations for neuron 4 - further analysis might be needed

the word "wouldn't" and its variations, indicating skepticism or hypothetical scenarios

No Comments

No Known Activations