INDEX

Explanations

dates or time frames

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "two" or other references to time spans

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 srfAttach

-0.73

angan

-0.72

MpServer

-0.67

Room

-0.64

Collect

-0.62

igham

-0.61

erella

-0.61

phot

-0.59

wrapper

-0.59

agall

-0.59

POSITIVE LOGITS

 iterations

0.85

 installments

0.85

 generations

0.83

imester

0.80

 weeks

0.79

days

0.79

encies

0.79

 millennium

0.76

cture

0.75

 semester

0.75

Activations Density 0.055%

dates or time frames

occurrences of the word "two" or other references to time spans

No Comments

No Known Activations