INDEX

Explanations

words related to conflict or intense situations, especially in the context of storytelling

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "drama" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ledge

-0.75

bin

-0.69

mun

-0.69

 Stores

-0.68

gans

-0.67

CDC

-0.66

initialized

-0.65

 Using

-0.64

Method

-0.64

 Enhanced

-0.64

POSITIVE LOGITS

 drama

3.67

 dramas

2.80

 Drama

2.60

rama

1.94

 thriller

1.83

 comedy

1.77

 suspense

1.57

 dram

1.52

 romance

1.49

 intrigue

1.42

Activations Density 0.017%

words related to conflict or intense situations, especially in the context of storytelling

mentions of the word "drama" and its variations

No Comments

No Known Activations