INDEX

Explanations

words related to riding or being on a journey

oai_token-act-pair · gpt-3.5-turbo

actions related to riding

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

aneously

-0.70

aneous

-0.69

essment

-0.68

âĶĢâĶĢ

-0.68

ident

-0.63

ayer

-0.62

razil

-0.62

ening

-0.61

arial

-0.61

 pronounced

-0.61

POSITIVE LOGITS

 motorcycles

1.03

horse

1.00

 bicycles

0.98

 bikes

0.96

 shotgun

0.95

bike

0.95

 horses

0.94

 horse

0.87

wagon

0.86

cycles

0.86

Activations Density 0.066%

words related to riding or being on a journey

actions related to riding

No Comments

No Known Activations