INDEX

Explanations

words related to revolutions or important events

oai_token-act-pair · gpt-3.5-turbo

references to significant historical events and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

Ire

-0.60

appropriate

-0.58

ĺ

-0.55

pport

-0.54

okÃ©

-0.54

ãĥ¼ãĥĨãĤ£

-0.53

ippy

-0.51

ł

-0.51

ullivan

-0.51

oyal

-0.51

POSITIVE LOGITS

 survives

1.14

 explodes

1.11

 occupies

1.08

 disappears

1.07

has

1.06

 belongs

1.06

 arrives

1.04

 enters

1.00

 consists

0.99

 awaits

0.99

Activations Density 0.671%

words related to revolutions or important events

references to significant historical events and entities

No Comments

No Known Activations