INDEX

Explanations

This neuron finds special characters used to segregate text

oai_token-act-pair · gpt-3.5-turbo

references to computer programming or technical concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hement

-0.92

 destro

-0.78

glers

-0.73

 nodd

-0.66

 ende

-0.66

NRS

-0.66

 estab

-0.65

 charism

-0.65

 Niet

-0.63

stad

-0.61

POSITIVE LOGITS

bus

0.82

shall

0.81

DATA

0.80

bable

0.79

INC

0.79

evidence

0.77

successful

0.77

FACE

0.76

heter

0.75

Activations Density 0.125%

This neuron finds special characters used to segregate text

references to computer programming or technical concepts

No Comments

No Known Activations