INDEX

Explanations

terms related to the conclusion or final part of a story

oai_token-act-pair · gpt-3.5-turbo

terms associated with the endings and finales of storytelling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

asus

-0.84

hee

-0.77

artisan

-0.75

enne

-0.74

uni

-0.72

ovi

-0.72

hammad

-0.72

Cola

-0.71

inen

-0.69

conn

-0.68

POSITIVE LOGITS

 endings

0.83

[|

0.73

 spoiler

0.72

 reel

0.70

 abruptly

0.67

blow

0.66

 wound

0.66

 prematurely

0.65

 tragically

0.65

 playthrough

0.64

Activations Density 0.073%

terms related to the conclusion or final part of a story

terms associated with the endings and finales of storytelling

No Comments

No Known Activations