INDEX

Explanations

terms related to neuroscience and medical research

oai_token-act-pair · gpt-3.5-turbo

concepts related to mechanisms of recognition and the role of personality in behavior

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

grim

-0.90

wered

-0.86

Rated

-0.82

renheit

-0.82

 scares

-0.76

Gall

-0.73

 Newsp

-0.70

pees

-0.70

 Bullets

-0.68

 Finish

-0.66

POSITIVE LOGITS

 mediated

1.29

 epigen

1.25

 mechanisms

1.21

 adaptive

1.19

 modulation

1.17

 medi

1.16

 pathways

1.08

 mechanism

1.07

mediated

1.04

 neuronal

1.04

Activations Density 0.434%

terms related to neuroscience and medical research

concepts related to mechanisms of recognition and the role of personality in behavior

No Comments

No Known Activations