INDEX

Explanations

terms related to information, data representation, and physical objects

oai_token-act-pair · gpt-3.5-turbo

concepts relating to information, objects, and data in various systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Veter

-0.74

udeb

-0.72

Brave

-0.70

atorium

-0.70

 Judd

-0.69

jah

-0.66

itone

-0.66

luaj

-0.66

CONCLUS

-0.65

 Nights

-0.65

POSITIVE LOGITS

 molecules

0.92

 behaving

0.91

 behaves

0.91

 intrinsically

0.91

 interacting

0.87

 whose

0.85

 undergoing

0.84

 encoded

0.83

 objects

0.83

 behave

0.83

Activations Density 0.425%

terms related to information, data representation, and physical objects

concepts relating to information, objects, and data in various systems

No Comments

No Known Activations