INDEX

Explanations

The neuron is looking for words related to "silence".

oai_token-act-pair · gpt-3.5-turbo

instances of a specific form of the word "sil"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

OTAL

-0.74

 Blessed

-0.68

 Ancients

-0.67

enegger

-0.67

 Briggs

-0.67

 damned

-0.66

Das

-0.62

 Herald

-0.62

 depreciation

-0.62

 Mother

-0.62

POSITIVE LOGITS

encers

1.44

encer

1.43

encing

1.40

enced

1.32

icone

1.23

ences

1.20

icate

1.14

icon

1.12

ica

1.02

entary

1.00

Activations Density 0.024%

**The neuron is looking for words related to "silence".**

instances of a specific form of the word "sil"

No Comments

No Known Activations

The neuron is looking for words related to "silence".