INDEX

Explanations

words related to achievements and changes in life circumstances

oai_token-act-pair · gpt-3.5-turbo

phrases indicating significant relationships or roles

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

 prescriptions

-0.76

 partitions

-0.70

 overlap

-0.67

 disag

-0.67

ãĤµ

-0.66

antage

-0.66

ousands

-0.66

 Explos

-0.66

 cancellation

-0.65

 indict

-0.64

POSITIVE LOGITS

 Alam

0.71

Sly

0.68

guard

0.66

coat

0.66

retty

0.66

mie

0.66

albeit

0.65

gets

0.65

 herself

0.64

puter

0.63

Activations Density 0.506%

words related to achievements and changes in life circumstances

phrases indicating significant relationships or roles

No Comments

No Known Activations