INDEX

Explanations

specific examples or instances mentioned within a broader context

oai_token-act-pair · gpt-3.5-turbo

mentions of examples

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

 mathemat

-0.88

emetery

-0.84

ternity

-0.83

vation

-0.82

iversal

-0.81

olulu

-0.78

orne

-0.78

kefeller

-0.77

 millenn

-0.75

ailability

-0.74

POSITIVE LOGITS

 CLSID

0.77

 Lily

0.74

 dummy

0.72

 subp

0.70

 code

0.70

 Example

0.69

 macros

0.68

Vag

0.66

Jas

0.65

Example

0.65

Activations Density 0.038%

specific examples or instances mentioned within a broader context

mentions of examples

No Comments

No Known Activations