INDEX

Explanations

references to attention and focus

oai_token-act-pair · gpt-3.5-turbo

terms related to attention and attention disorders

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Mehran

-0.70

 halves

-0.67

 Tale

-0.66

tto

-0.65

 Recon

-0.64

ourke

-0.62

 Yugoslavia

-0.62

 Castro

-0.61

raine

-0.61

 Dani

-0.60

POSITIVE LOGITS

estinal

1.00

orial

0.86

 attention

0.81

ively

0.80

arios

0.77

stadt

0.76

 largeDownload

0.74

cipline

0.74

 spans

0.73

flow

0.73

Activations Density 0.017%

references to attention and focus

terms related to attention and attention disorders

No Comments

No Known Activations