INDEX

Explanations

technical terms or concepts related to approaches or strategies

oai_token-act-pair · gpt-3.5-turbo

terms related to strategies, approaches, and conceptual frameworks

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

farious

-0.62

reporting

-0.57

CLR

-0.56

ergy

-0.56

sci

-0.55

charging

-0.54

this

-0.54

--------------------------------------------------------

-0.54

 Right

-0.53

Latest

-0.53

POSITIVE LOGITS

 coincides

1.10

 extends

1.09

 represents

1.08

 ensures

1.05

 encompasses

1.04

 belongs

1.03

 applies

1.03

 translates

1.03

 assumes

1.02

 corresponds

1.02

Activations Density 0.257%

technical terms or concepts related to approaches or strategies

terms related to strategies, approaches, and conceptual frameworks

No Comments

No Known Activations