INDEX

Explanations

individualized statements or perspectives

oai_token-act-pair · gpt-3.5-turbo

connections or relationships between subjects or entities

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

":-

-0.67

(?,

-0.65

hess

-0.65

ourse

-0.64

 sclerosis

-0.63

malink

-0.63

-0.61

usercontent

-0.60

 Nare

-0.59

enes

-0.58

POSITIVE LOGITS

arently

0.88

ardless

0.76

)</

0.70

ĪĴ

0.69

ãĥ¼ãĥĨãĤ£

0.66

-)

0.66

ãĥĭ

0.64

itored

0.63

 incidentally

0.63

ãĤ¨ãĥ«

0.63

Activations Density 0.274%

individualized statements or perspectives

connections or relationships between subjects or entities

No Comments

No Known Activations