INDEX

Explanations

scientific research terms and methodologies

oai_token-act-pair · gpt-3.5-turbo

key findings and results from scientific investigations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Armageddon

-0.78

ById

-0.78

atron

-0.73

soDeliveryDate

-0.71

Kids

-0.69

shit

-0.68

 Slayer

-0.67

Maker

-0.65

 Bohem

-0.65

 crap

-0.63

POSITIVE LOGITS

 analyses

1.06

 hypothesized

1.04

RESULTS

1.04

 hypotheses

1.00

Methods

0.99

Materials

0.99

 phylogen

0.96

 systematic

0.94

ivariate

0.93

 herein

0.92

Activations Density 0.282%

scientific research terms and methodologies

key findings and results from scientific investigations

No Comments

No Known Activations