INDEX

Explanations

terms related to experiments and research

oai_token-act-pair · gpt-3.5-turbo

mentions of experiments

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

headers

-0.66

 grievances

-0.63

cut

-0.61

doms

-0.60

clinton

-0.60

roots

-0.59

ĺħ

-0.59

BuyableInstoreAndOnline

-0.58

othy

-0.58

entity

-0.57

POSITIVE LOGITS

ally

1.15

imental

1.05

iments

0.99

ationally

0.90

 experiment

0.90

 Experiment

0.88

ary

0.84

Exper

0.83

ations

0.83

ivity

0.82

Activations Density 0.042%

terms related to experiments and research

mentions of experiments

No Comments

No Known Activations