INDEX

Explanations

phrases related to addition or inclusion

oai_token-act-pair · gpt-3.5-turbo

references to additional information or considerations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 QUEST

-0.66

pload

-0.63

resa

-0.61

ammy

-0.61

 Tray

-0.60

 Nurs

-0.60

 RAID

-0.59

itialized

-0.59

 meanwhile

-0.56

 moreover

-0.56

POSITIVE LOGITS

.</

0.74

ãĤ´ãĥ³

0.72

.",

0.71

ãĢĤ

0.70

.[

0.70

.ï¿½

0.68

0.67

because

0.65

;)

0.64

 guiActiveUnfocused

0.63

Activations Density 0.193%

phrases related to addition or inclusion

references to additional information or considerations

No Comments

No Known Activations