INDEX

Explanations

phrases where something is being put together or compiled

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "put together."

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

den

-0.70

null

-0.68

outer

-0.67

GA

-0.63

DEN

-0.62

izu

-0.61

esa

-0.61

externalActionCode

-0.60

azed

-0.60

jin

-0.60

POSITIVE LOGITS

ãĤ¤ãĥĪ

0.89

sonian

0.78

illet

0.74

itions

0.74

acular

0.74

igham

0.74

ilts

0.71

uggest

0.69

ederation

0.69

éĹĺ

0.66

Activations Density 0.016%

phrases where something is being put together or compiled

instances of the phrase "put together."

No Comments

No Known Activations