INDEX

Explanations

references to the year 2006

oai_token-act-pair · gpt-3.5-turbo

occurrences of the year 2006 in various contexts

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

ymes

-0.75

otions

-0.69

amins

-0.66

plom

-0.65

 lawy

-0.63

ances

-0.62

 Vader

-0.61

 ancest

-0.60

friends

-0.60

 starship

-0.60

POSITIVE LOGITS

å¹

0.95

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.72

 vintage

0.71

velength

0.66

 onwards

0.65

Fuk

0.64

thritis

0.64

cture

0.63

 Springs

0.63

0.62

Activations Density 0.021%

references to the year 2006

occurrences of the year 2006 in various contexts

No Comments

No Known Activations