INDEX

Explanations

references to the origin or source of something

oai_token-act-pair · gpt-3.5-turbo

phrases relating to the origin of various entities or concepts

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

vil

-0.82

thur

-0.75

owl

-0.73

istors

-0.70

err

-0.70

aving

-0.70

nature

-0.70

eely

-0.70

evaluate

-0.69

dain

-0.66

POSITIVE LOGITS

REDACTED

0.85

 originating

0.76

Ú

0.75

ATED

0.75

 originate

0.74

ially

0.69

ATING

0.67

 originated

0.64

ãĥ¼ãĥĨãĤ£

0.64

ators

0.63

Activations Density 0.017%

references to the origin or source of something

phrases relating to the origin of various entities or concepts

No Comments

No Known Activations