INDEX

Explanations

specific details related to different physical objects or items, like modifications, components, materials, or accessories

oai_token-act-pair · gpt-3.5-turbo

commas in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

haus

-0.67

ovich

-0.66

 Chapters

-0.66

²¾

-0.65

 Equity

-0.65

itate

-0.64

 Individuals

-0.62

 Investors

-0.62

pace

-0.61

endo

-0.61

POSITIVE LOGITS

 which

1.09

aka

0.97

which

0.92

 nicknamed

0.88

Magikarp

0.78

 wherein

0.77

 whereby

0.77

 consisting

0.75

 mustache

0.73

Downloadha

0.73

Activations Density 0.367%

specific details related to different physical objects or items, like modifications, components, materials, or accessories

commas in the text

No Comments

No Known Activations