INDEX

Explanations

questions related to the presence of specific things or actions

oai_token-act-pair · gpt-3.5-turbo

questions and conditional statements about various topics

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

burgh

-0.64

loo

-0.63

haw

-0.62

pes

-0.58

onomy

-0.57

cius

-0.56

loving

-0.55

staking

-0.55

mostly

-0.54

 srfAttach

-0.54

POSITIVE LOGITS

any

2.24

 anything

2.12

ANY

1.84

 anybody

1.76

anything

1.74

 anyone

1.71

 anywhere

1.57

any

1.53

Any

1.50

Any

1.46

Activations Density 0.765%

questions related to the presence of specific things or actions

questions and conditional statements about various topics

No Comments

No Known Activations