INDEX

Explanations

self-referential statements or commands

oai_token-act-pair · gpt-3.5-turbo

statements that convey conditionality, specifically beginning with "unless."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Topics

-0.79

listed

-0.65

ãĥ´

-0.64

ä¹ĭ

-0.63

"]=>

-0.63

rift

-0.61

âĢİ

-0.61

 Drawn

-0.58

favorite

-0.58

alde

-0.58

POSITIVE LOGITS

 somehow

0.81

 expressly

0.78

 willfully

0.74

 explicitly

0.74

ispers

0.72

 specifically

0.69

 medically

0.66

 disguise

0.65

 magically

0.65

 manually

0.64

Activations Density 0.139%

self-referential statements or commands

statements that convey conditionality, specifically beginning with "unless."

No Comments

No Known Activations