INDEX

Explanations

terms related to comments sections, including phrases like "Comments welcome" and variations of "Comments #" and "posted"

oai_token-act-pair · gpt-3.5-turbo

words related to comments and feedback sections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

planes

-0.76

plane

-0.67

ties

-0.66

ously

-0.66

 faces

-0.66

aways

-0.65

yards

-0.64

eur

-0.62

ALLY

-0.62

 mates

-0.62

POSITIVE LOGITS

ource

1.18

ystem

1.03

cript

1.01

poons

1.00

ourcing

0.98

heet

0.97

pace

0.97

pring

0.97

uggest

0.97

ettings

0.97

Activations Density 0.155%

terms related to comments sections, including phrases like "Comments welcome" and variations of "Comments #" and "posted"

words related to comments and feedback sections

No Comments

No Known Activations