INDEX

Explanations

verbs followed by pronouns indicating actions or consequences that may happen as a result of a specific behavior

oai_token-act-pair · gpt-3.5-turbo

references to the pronoun "you" in various contexts

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

ãĤ´ãĥ³

-0.79

surprisingly

-0.71

ãĤ¤ãĥĪ

-0.65

 Estimates

-0.64

Print

-0.62

albeit

-0.62

ULTS

-0.62

 Anchorage

-0.62

Assembly

-0.62

 âĵĺ

-0.61

POSITIVE LOGITS

 wanna

1.26

're

1.23

ain

1.08

 criticize

0.99

 lose

0.98

 offend

0.97

 injure

0.94

 succeed

0.94

 want

0.94

 piss

0.94

Activations Density 0.170%

verbs followed by pronouns indicating actions or consequences that may happen as a result of a specific behavior

references to the pronoun "you" in various contexts

No Comments

No Known Activations