INDEX

Explanations

terms related to data analysis and examination

oai_token-act-pair · gpt-3.5-turbo

instances of the word "analyze" 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

orthy

-0.68

orld

-0.67

bearing

-0.67

eworld

-0.65

vic

-0.64

bringing

-0.64

cot

-0.63

Lago

-0.63

birth

-0.63

arat

-0.62

POSITIVE LOGITS

 analy

1.08

 analyze

1.07

analy

0.98

 analyzing

0.97

 partName

0.85

 Analy

0.85

racuse

0.85

 analys

0.84

 Interpret

0.83

rador

0.82

Activations Density 0.013%

terms related to data analysis and examination

instances of the word "analyze" and its variations

No Comments

No Known Activations