INDEX

Explanations

phrases that express a strong opinion or make a judgment

oai_token-act-pair · gpt-3.5-turbo

phrases that denote a condition or situation described with the word "that."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Voting

-0.70

anton

-0.69

SD

-0.66

 Aden

-0.65

known

-0.64

 Modified

-0.63

episode

-0.62

ESE

-0.61

classified

-0.61

Previous

-0.61

POSITIVE LOGITS

 deserves

1.16

 threatens

1.11

 justifies

1.09

 undermines

1.07

 attracts

1.07

 overwhel

1.07

 eats

1.06

 refuses

1.05

 lacks

1.04

 awaits

1.03

Activations Density 0.269%

phrases that express a strong opinion or make a judgment

phrases that denote a condition or situation described with the word "that."

No Comments

No Known Activations