INDEX

Explanations

tense words often used in reviews, such as "it is", "could be", "should work"

oai_token-act-pair · gpt-3.5-turbo

references to challenges and complexities in processes or situations

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

 Facts

-0.71

hovah

-0.67

ylum

-0.67

 socialist

-0.63

 Communist

-0.62

 Rodham

-0.61

 Government

-0.59

 Nazi

-0.59

 Socialist

-0.59

igion

-0.59

POSITIVE LOGITS

 noticeably

0.86

 offset

0.85

 noticeable

0.82

 smoother

0.81

 slower

0.81

 offsets

0.79

 compensate

0.77

 downgrade

0.75

 optimizations

0.75

 tweaks

0.75

Activations Density 1.007%

tense words often used in reviews, such as "it is", "could be", "should work"

references to challenges and complexities in processes or situations

No Comments

No Known Activations