INDEX

Explanations

instances of the word "laser" with a significant activation level

oai_token-act-pair · gpt-3.5-turbo

references to lasers and their applications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

fam

-0.85

skirts

-0.78

 Alive

-0.76

gres

-0.76

itutional

-0.75

ISTORY

-0.72

 Chatt

-0.72

ãģį

-0.70

annah

-0.69

 Rough

-0.69

POSITIVE LOGITS

 beam

1.02

 pointer

1.00

 beams

0.91

 cutter

0.91

 cannons

0.85

 laser

0.85

 projector

0.84

ray

0.83

 lasers

0.82

 cannon

0.82

Activations Density 0.026%

instances of the word "laser" with a significant activation level

references to lasers and their applications

No Comments

No Known Activations