INDEX

Explanations

numbers referring to quantities or attributes of items in a list

oai_token-act-pair · gpt-3.5-turbo

numerical values or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Aly

-0.70

amaru

-0.66

RTX

-0.62

 jerk

-0.61

plin

-0.60

Rex

-0.59

rection

-0.57

aeus

-0.57

imaru

-0.57

 derog

-0.57

POSITIVE LOGITS

th

1.09

0.93

teen

0.92

Â½

0.91

 million

0.89

0.88

0.87

Activations Density 0.279%

numbers referring to quantities or attributes of items in a list

numerical values or statistics

No Comments

No Known Activations