INDEX

Explanations

numbers, such as amounts, measurements, and identifiers

oai_token-act-pair · gpt-3.5-turbo

numerical data and figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Buff

-0.74

Cap

-0.69

 Magic

-0.68

Col

-0.68

-0.67

Big

-0.66

Ta

-0.64

 Fort

-0.63

 Final

-0.63

 Week

-0.63

POSITIVE LOGITS

2.15

2.13

2.12

2.11

2.10

2.09

Activations Density 0.034%

numbers, such as amounts, measurements, and identifiers

numerical data and figures

No Comments

No Known Activations