INDEX

Explanations

terms related to hierarchical structures, specifically focusing on the concept of a 'tree'

oai_token-act-pair · gpt-3.5-turbo

references to various types of trees within a technical or metaphorical context

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

glomer

-0.99

ione

-0.79

abwe

-0.78

rontal

-0.78

arcer

-0.78

aleigh

-0.77

ities

-0.77

ividual

-0.75

farious

-0.74

atility

-0.73

POSITIVE LOGITS

frog

1.08

 canopy

1.01

beard

1.00

 Tree

1.00

 tree

0.97

 stump

0.97

 trunk

0.95

 trees

0.95

Hug

0.92

yard

0.92

Activations Density 0.014%

terms related to hierarchical structures, specifically focusing on the concept of a 'tree'

references to various types of trees within a technical or metaphorical context

No Comments

No Known Activations