INDEX

Explanations

terms related to armor and weaponry

oai_token-act-pair · gpt-3.5-turbo

terms related to armor and protective gear

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zee

-0.78

 Lakes

-0.76

liam

-0.72

 Vide

-0.67

istani

-0.66

arak

-0.65

Bee

-0.63

Suc

-0.63

ITNESS

-0.62

Gos

-0.61

POSITIVE LOGITS

 piercing

1.21

 Piercing

1.13

ments

1.09

 clad

1.01

 armour

0.99

 thickness

0.99

 worn

0.97

 armor

0.94

clad

0.93

armor

0.90

Activations Density 0.017%

terms related to armor and weaponry

terms related to armor and protective gear

No Comments

No Known Activations