INDEX

Explanations

terms related to exploring or investigating various topics or places

oai_token-act-pair · gpt-3.5-turbo

instances of the word "exploring" and its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 perjury

-0.78

 phony

-0.68

 Lump

-0.67

raud

-0.67

 hypoc

-0.67

 defective

-0.64

 Fraud

-0.64

fax

-0.63

 Customer

-0.63

çļ

-0.63

POSITIVE LOGITS

 exploring

3.14

 explore

2.61

 explored

2.61

 exploration

2.23

 explores

2.08

 explor

2.02

 Explore

1.69

Explore

1.63

 Exploration

1.58

 explorers

1.54

Activations Density 0.030%

terms related to exploring or investigating various topics or places

instances of the word "exploring" and its variations

No Comments

No Known Activations