INDEX

Explanations

the phrase "no one" followed by a verb or verb phrase, indicating actions or situations

oai_token-act-pair · gpt-3.5-turbo

the repeated use of the word "one" 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 @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

éŃĶ

-0.80

tnc

-0.71

YN

-0.70

ailability

-0.61

 exting

-0.61

ORK

-0.61

å½

-0.61

ò

-0.60

srf

-0.60

rus

-0.60

POSITIVE LOGITS

 else

1.43

 cares

1.29

 knows

1.27

 bothers

1.19

 expects

1.13

 wants

1.10

 understands

1.09

 remembers

1.05

 cared

1.03

 likes

1.03

Activations Density 0.064%

the phrase "no one" followed by a verb or verb phrase, indicating actions or situations

the repeated use of the word "one" in various contexts

No Comments

No Known Activations