INDEX

Explanations

adjectives or participles as prefixes in the text

oai_token-act-pair · gpt-3.5-turbo

negations or prefixes indicating a lack or absence of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 blasts

-0.80

 disclaimer

-0.71

 brushes

-0.69

 Revelations

-0.68

 pics

-0.68

 Anniversary

-0.68

 understatement

-0.67

 tones

-0.66

 comma

-0.66

 sake

-0.66

POSITIVE LOGITS

responsive

1.62

usable

1.28

productive

1.28

accessible

1.22

functional

1.18

interested

1.17

aggressive

1.16

compl

1.15

threatening

1.13

aware

1.12

Activations Density 0.098%

adjectives or participles as prefixes in the text

negations or prefixes indicating a lack or absence of something

No Comments

No Known Activations