INDEX

Explanations

references to neurons in brain-related contexts

oai_token-act-pair · gpt-3.5-turbo

references to neurons and their characteristics

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

 Vend

-0.85

aid

-0.77

weather

-0.72

edd

-0.72

 Carnival

-0.71

styles

-0.70

irl

-0.70

tar

-0.69

 Destruction

-0.65

bris

-0.65

POSITIVE LOGITS

 neurons

3.47

 neuron

3.05

 neuronal

2.27

 hippocamp

1.71

 cortical

1.69

 cortex

1.67

 synaptic

1.64

 neural

1.58

 hippocampus

1.52

 dopamine

1.50

Activations Density 0.033%

references to neurons in brain-related contexts

references to neurons and their characteristics

No Comments

No Known Activations