INDEX

Explanations

references to various items or concepts that are not explicitly mentioned in the document

oai_token-act-pair · gpt-3.5-turbo

the term "other" 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

-0.81

 Kubrick

-0.69

utenant

-0.67

ctor

-0.65

 Cycling

-0.64

 Jaguar

-0.63

ister

-0.63

 Potion

-0.63

POSITIVE LOGITS

worldly

1.54

 assorted

1.11

 aspects

0.90

 kinds

0.90

 types

0.90

 facets

0.88

 goodies

0.85

 forms

0.84

quickShipAvailable

0.82

 indications

0.81

Activations Density 0.063%

references to various items or concepts that are not explicitly mentioned in the document

the term "other" in various contexts

No Comments

No Known Activations