INDEX

Explanations

dates and time periods

oai_token-act-pair · gpt-3.5-turbo

phrases indicating date ranges or time periods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

natureconservancy

-0.72

 Zerg

-0.67

inct

-0.62

 welf

-0.62

 blat

-0.61

utenberg

-0.60

 orcs

-0.59

 deviations

-0.58

ocus

-0.58

ingu

-0.58

POSITIVE LOGITS

 October

1.46

 July

1.43

 September

1.43

 June

1.43

 December

1.41

 April

1.39

 August

1.38

 November

1.35

 March

1.34

 February

1.33

Activations Density 0.048%

dates and time periods

phrases indicating date ranges or time periods

No Comments

No Known Activations