INDEX

Explanations

statements related to political leadership and government officials

oai_token-act-pair · gpt-3.5-turbo

possesive forms of nouns, particularly indicating ownership or association

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ording

-0.86

nels

-0.78

iator

-0.76

subject

-0.73

andowski

-0.72

oyer

-0.71

Subject

-0.70

ucl

-0.70

object

-0.69

!/

-0.69

POSITIVE LOGITS

 largest

1.17

 finest

1.02

 wealthiest

1.00

 richest

1.00

 newest

0.99

 poorest

0.99

 tallest

0.98

 biggest

0.97

 oldest

0.97

 premier

0.96

Activations Density 0.112%

statements related to political leadership and government officials

possesive forms of nouns, particularly indicating ownership or association

No Comments

No Known Activations