INDEX

Explanations

words related to brain or mental activity

oai_token-act-pair · gpt-3.5-turbo

mentions of the brain and its various conditions or activities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

leased

-0.77

ilts

-0.73

atives

-0.72

ime

-0.71

iquette

-0.67

ival

-0.67

âĢİ

-0.66

wind

-0.66

clus

-0.66

 Holder

-0.64

POSITIVE LOGITS

 brain

3.47

 brains

2.78

brain

2.71

Brain

2.47

 Brain

2.43

 cerebral

1.99

 prefrontal

1.80

 neural

1.76

 neuro

1.76

 neurological

1.73

Activations Density 0.026%

words related to brain or mental activity

mentions of the brain and its various conditions or activities

No Comments

No Known Activations