INDEX

Explanations

terms related to fortified or strongly defended locations

oai_token-act-pair · gpt-3.5-turbo

terms related to strongholds or places of power

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

hy

-0.86

matter

-0.86

ivan

-0.81

ensation

-0.79

ordan

-0.77

andro

-0.76

kay

-0.76

respond

-0.75

crew

-0.74

videos

-0.72

POSITIVE LOGITS

 stronghold

1.26

 strongh

1.19

 bast

0.82

 elector

0.81

urst

0.80

 advoc

0.78

 ende

0.71

 hinges

0.71

 adherent

0.70

 indo

0.69

Activations Density 0.015%

terms related to fortified or strongly defended locations

terms related to strongholds or places of power

No Comments

No Known Activations