INDEX

Explanations

instances where there are multiple similar nouns or names mentioned consecutively

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "there"

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

 externalToEVAOnly

-0.60

 Submit

-0.59

 Achieve

-0.59

 ILCS

-0.59

 bestselling

-0.56

 Sending

-0.56

 Doing

-0.55

 Deliver

-0.55

 Transfer

-0.55

 Warfare

-0.54

POSITIVE LOGITS

 were

1.51

 weren

1.51

 wasn

1.38

was

1.36

upon

1.29

 seemed

1.25

were

1.15

 hadn

1.15

 appeared

1.13

WAS

1.12

Activations Density 0.075%

instances where there are multiple similar nouns or names mentioned consecutively

the presence of the word "there"

No Comments

No Known Activations