INDEX

Explanations

phrases or sentences mentioning that someone has made a comment or expressed an opinion

oai_token-act-pair · gpt-3.5-turbo

the phrase "have" indicating ongoing actions or states

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Achieve

-0.75

 Serve

-0.69

 Kills

-0.67

eem

-0.65

 Provide

-0.65

 Deliver

-0.64

 fulfil

-0.64

 srfAttach

-0.63

 Enhance

-0.62

 Expand

-0.60

POSITIVE LOGITS

 speculated

1.30

 wondered

1.25

 begun

1.17

 been

1.14

 complained

1.13

 argued

1.08

 reacted

1.07

 questioned

1.07

 noticed

1.07

 rallied

1.06

Activations Density 0.130%

phrases or sentences mentioning that someone has made a comment or expressed an opinion

the phrase "have" indicating ongoing actions or states

No Comments

No Known Activations