INDEX

Explanations

action verbs related to problem-solving or analysis

oai_token-act-pair · gpt-3.5-turbo

elements related to analysis and evaluation in a given context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 helmets

-0.76

 Empires

-0.69

 brokers

-0.67

 mosques

-0.66

 cannons

-0.65

 jails

-0.64

 shotguns

-0.64

 Poles

-0.62

 Borders

-0.61

breeding

-0.60

POSITIVE LOGITS

something

0.89

or

0.85

 stimuli

0.82

Something

0.81

etc

0.81

 satisf

0.80

terday

0.79

Previous

0.79

(*

0.78

Completed

0.78

Activations Density 0.624%

action verbs related to problem-solving or analysis

elements related to analysis and evaluation in a given context

No Comments

No Known Activations