INDEX

Explanations

references to specific quantities or amounts, especially in relation to materials like grain or emissions

oai_token-act-pair · gpt-3.5-turbo

mentions of quantities measured in tonnes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

comes

-0.74

States

-0.73

states

-0.70

Rap

-0.63

gotten

-0.63

 Presidents

-0.62

Leader

-0.62

Pir

-0.60

leader

-0.60

hall

-0.60

POSITIVE LOGITS

 tonnes

1.17

atsu

0.93

chwitz

0.86

lbs

0.84

 litres

0.81

 greenhouse

0.79

ascus

0.79

 stockp

0.78

 loads

0.78

arge

0.78

Activations Density 0.010%

references to specific quantities or amounts, especially in relation to materials like grain or emissions

mentions of quantities measured in tonnes

No Comments

No Known Activations