INDEX

Explanations

events or instances that happened for the last time

oai_token-act-pair · gpt-3.5-turbo

instances of specific notable events or turning points in timelines

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

astical

-0.78

ciples

-0.75

complex

-0.73

 abound

-0.72

igm

-0.71

ogle

-0.69

ampunk

-0.68

aintain

-0.67

ims

-0.66

orian

-0.65

POSITIVE LOGITS

 February

0.87

 September

0.86

 December

0.86

 November

0.82

0.80

 October

0.79

 January

0.79

 April

0.75

 March

0.75

Activations Density 0.705%

events or instances that happened for the last time

instances of specific notable events or turning points in timelines

No Comments

No Known Activations