INDEX

Explanations

terms related to theoretical concepts or hypotheses

oai_token-act-pair · gpt-3.5-turbo

references to various scientific theories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Pradesh

-0.68

 Mous

-0.67

 Samar

-0.67

ãģį

-0.65

iences

-0.64

ienced

-0.63

ments

-0.63

å¸

-0.60

iencies

-0.60

Delivery

-0.60

POSITIVE LOGITS

icist

1.05

ologies

1.02

 underpin

1.01

craft

0.89

sonian

0.89

 theories

0.82

ical

0.81

 theor

0.80

 physicist

0.79

 predicts

0.76

Activations Density 0.055%

terms related to theoretical concepts or hypotheses

references to various scientific theories

No Comments

No Known Activations