INDEX

Explanations

phrases related to the term "im" followed by various letters or numbers

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "Im" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

NetMessage

-0.90

 Nazis

-0.78

 grapes

-0.75

 mars

-0.74

 wool

-0.72

 shroud

-0.71

 Skydragon

-0.71

 Chrys

-0.70

yards

-0.69

ttes

-0.69

POSITIVE LOGITS

ple

1.38

mediate

1.37

mediately

1.37

perfect

1.26

perial

1.26

agine

1.19

unity

1.19

posing

1.18

ply

1.17

plementation

1.17

Activations Density 0.010%

phrases related to the term "im" followed by various letters or numbers

the presence of the word "Im" in various contexts

No Comments

No Known Activations