INDEX

Explanations

scientific experiments and tests

oai_token-act-pair · gpt-3.5-turbo

terms related to scientific testing and validation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Destination

-0.71

 nods

-0.68

resa

-0.65

interstitial

-0.65

 holiest

-0.61

 emphasis

-0.61

 announcements

-0.61

edo

-0.60

_-

-0.59

 blame

-0.59

POSITIVE LOGITS

 commercially

1.13

 empir

1.11

 reliably

1.06

 experiment

1.04

 using

1.03

 indoors

1.03

 manually

1.02

 scientifically

1.01

 microsc

0.96

 successfully

0.95

Activations Density 0.352%

scientific experiments and tests

terms related to scientific testing and validation

No Comments

No Known Activations