INDEX

Explanations

phrases related to underlying issues or structures

oai_token-act-pair · gpt-3.5-turbo

the concept of underlying factors or issues

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

ooters

-0.88

asia

-0.87

cture

-0.81

nery

-0.76

ishers

-0.75

kie

-0.73

itto

-0.72

efully

-0.71

oping

-0.71

cker

-0.70

POSITIVE LOGITS

 assumptions

0.87

 underlying

0.86

 motivations

0.84

 infrastructure

0.84

 fundamentals

0.83

 principles

0.82

 underpin

0.80

 layers

0.79

 structures

0.78

 assumption

0.77

Activations Density 0.015%

phrases related to underlying issues or structures

the concept of underlying factors or issues

No Comments

No Known Activations