INDEX

Explanations

phrases that express a general assessment or summary of a situation or topic

oai_token-act-pair · gpt-3.5-turbo

references to overall assessments or evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

DragonMagazine

-0.77

Mini

-0.70

 Secondly

-0.70

©¶æ

-0.70

cells

-0.67

Caf

-0.66

rosso

-0.66

Secondly

-0.64

 Reborn

-0.64

Fax

-0.64

POSITIVE LOGITS

 impression

0.94

 picture

0.93

 attractiveness

0.87

 satisfaction

0.84

 wellbeing

0.84

 tone

0.82

 length

0.81

 outlook

0.81

 direction

0.80

 standings

0.80

Activations Density 0.042%

phrases that express a general assessment or summary of a situation or topic

references to overall assessments or evaluations

No Comments

No Known Activations