INDEX

Explanations

dates and numerical values structured in a specific way

oai_token-act-pair · gpt-3.5-turbo

terms related to classifications and categories within various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Wan

-0.90

mosp

-0.89

hire

-0.84

Kick

-0.81

 Peng

-0.77

Mir

-0.77

ffe

-0.76

 Wellington

-0.73

 Faith

-0.73

Lee

-0.71

POSITIVE LOGITS

 classification

2.08

 category

1.89

 categories

1.86

 categor

1.81

 classify

1.78

 Classification

1.62

 Categories

1.58

Category

1.54

 subclass

1.54

 Category

1.53

Activations Density 0.400%

dates and numerical values structured in a specific way

terms related to classifications and categories within various contexts

No Known Activations