INDEX

Explanations

language indicating future expectations or possibilities

oai_token-act-pair · gpt-3.5-turbo

phrases expressing future expectations or predictions

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

 sincerity

-0.66

PLIED

-0.63

 Contributions

-0.61

rice

-0.61

 volunt

-0.60

forms

-0.59

 teammates

-0.59

 personality

-0.59

 charisma

-0.58

 vanity

-0.58

POSITIVE LOGITS

 witnessing

1.28

 glimpse

1.07

 witness

1.05

 seeing

1.03

 glimps

1.01

 witnessed

0.98

 hear

0.98

 hearing

0.92

 ourselves

0.92

see

0.90

Activations Density 0.310%

language indicating future expectations or possibilities

phrases expressing future expectations or predictions

No Comments

No Known Activations