INDEX

Explanations

mentions of tents

oai_token-act-pair · gpt-3.5-turbo

mentions of tents

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

bye

-0.68

Words

-0.64

 Cuomo

-0.63

History

-0.61

 symmetry

-0.60

Finish

-0.60

obbies

-0.60

enforcement

-0.59

period

-0.59

sen

-0.59

POSITIVE LOGITS

acle

1.22

pole

1.22

 tent

1.18

atively

1.15

 encamp

1.05

 tents

1.03

 pole

0.98

acles

0.95

 poles

0.87

 flap

0.86

Activations Density 0.009%

mentions of tents

mentions of tents

No Comments

No Known Activations