INDEX

Explanations

The neuron identifies capital letters followed by symbols representing mathematical actions

oai_token-act-pair · gpt-3.5-turbo

specific symbols or characters that are not easily categorized

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

front

-0.66

 whit

-0.63

 Pupp

-0.61

ish

-0.60

bug

-0.60

 camouflage

-0.60

sm

-0.59

 Virgin

-0.58

face

-0.58

 doom

-0.57

POSITIVE LOGITS

Ģ

4.42

ģ

2.07

ĥ

1.99

Ħ

1.99

Ī

1.86

Ĥ

1.85

ī

1.76

ķ

1.75

į

1.75

Į

1.72

Activations Density 0.011%

The neuron identifies capital letters followed by symbols representing mathematical actions

specific symbols or characters that are not easily categorized

No Comments

No Known Activations