INDEX

Explanations

references to cages

oai_token-act-pair · gpt-3.5-turbo

mentions of "cage" and its related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Grad

-0.72

Hur

-0.70

properties

-0.69

appropri

-0.68

hur

-0.67

 Redd

-0.67

MLA

-0.66

 Relief

-0.66

lly

-0.65

Kot

-0.64

POSITIVE LOGITS

 cage

3.90

 cages

3.09

 Cage

1.61

 enclosure

1.58

 fence

1.28

 crate

1.19

 fences

1.08

 leash

1.07

 feather

1.03

 compartment

1.01

Activations Density 0.011%

references to cages

mentions of "cage" and its related concepts

No Comments

No Known Activations