INDEX

Explanations

scientific or technical terms related to stimuli or causes of certain responses

oai_token-act-pair · gpt-3.5-turbo

terms related to media, technology, and social commentary

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 unden

-0.53

ULL

-0.51

:,

-0.51

senal

-0.49

addon

-0.49

 Curious

-0.47

+.

-0.47

GGGGGGGG

-0.47

 Invention

-0.46

 Solution

-0.45

POSITIVE LOGITS

 cannot

0.78

 could

0.78

 lacks

0.76

 exists

0.75

 existed

0.74

 would

0.73

had

0.72

 should

0.72

 might

0.71

 ought

0.71

Activations Density 0.937%

scientific or technical terms related to stimuli or causes of certain responses

terms related to media, technology, and social commentary

No Comments

No Known Activations