INDEX

Explanations

direct speech with quotations

oai_token-act-pair · gpt-3.5-turbo

quotation marks indicating spoken or quoted dialogue

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

 prec

-0.82

 prelim

-0.78

 pacing

-0.77

 altar

-0.73

 halfway

-0.72

 streaming

-0.72

 priesthood

-0.72

 studio

-0.70

 ranking

-0.70

 pyramid

-0.70

POSITIVE LOGITS

Really

1.20

We

1.17

quickShipAvailable

1.15

Certainly

1.14

1.12

What

1.11

Very

1.10

Quite

1.09

You

1.09

It

1.09

Activations Density 0.096%

direct speech with quotations

quotation marks indicating spoken or quoted dialogue

No Comments

No Known Activations