INDEX

Explanations

specific references to 'the' followed by a significant noun or concept

oai_token-act-pair · gpt-3.5-turbo

the frequent use of the word "the" in contexts describing significant issues or events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atures

-1.00

classes

-0.94

agues

-0.93

hops

-0.91

events

-0.90

endars

-0.86

devices

-0.84

agents

-0.84

idents

-0.84

oots

-0.83

POSITIVE LOGITS

 fact

1.34

 insistence

1.31

 inability

1.31

 realization

1.28

 notion

1.23

 tendency

1.20

 ability

1.20

 assumption

1.19

 willingness

1.18

 belief

1.17

Activations Density 0.229%

specific references to 'the' followed by a significant noun or concept

the frequent use of the word "the" in contexts describing significant issues or events

No Comments

No Known Activations