INDEX

Explanations

terms related to computer coding or programming

oai_token-act-pair · gpt-3.5-turbo

references to specific projects or productions within a larger context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

hement

-0.88

 agre

-0.78

 compan

-0.77

bley

-0.72

 competitiveness

-0.70

 acknow

-0.69

 concess

-0.68

undai

-0.68

 disadvant

-0.66

 colleg

-0.66

POSITIVE LOGITS

Offline

1.16

Legendary

1.10

Premium

0.98

 Reloaded

0.88

join

0.86

Adapt

0.85

Joined

0.84

minster

0.80

Hero

0.78

Main

0.76

Activations Density 0.208%

terms related to computer coding or programming

references to specific projects or productions within a larger context

No Comments

No Known Activations