INDEX

Explanations

references to specific people or characters

oai_token-act-pair · gpt-3.5-turbo

references to emotional connections and relationships

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

Dob

-0.85

 sail

-0.79

 torpedo

-0.77

 Gron

-0.77

 FreeBSD

-0.76

 Newport

-0.75

 Berm

-0.74

NEO

-0.74

 Moons

-0.73

 Nelson

-0.73

POSITIVE LOGITS

âĢ

2.18

 âĢ

1.47

ãĢ

1.19

âĢł

1.10

Â»

1.09

*.

1.07

âĸł

1.04

_.

1.02

âĻ

1.00

________________________________

0.99

Activations Density 1.277%

references to specific people or characters

references to emotional connections and relationships

No Comments

No Known Activations