INDEX

Explanations

the word "from."

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

****************************************************************

-1.44

 Channel

-1.44

 welcome

-1.43

ty

-1.42

rvert

-1.40

 nick

-1.34

OUS

-1.31

âĢł

-1.30

"}**).

-1.30

OME

-1.30

POSITIVE LOGITS

ĻĤ

3.54

Į

3.51

¼

3.27

ľĵ

3.25

↵↵

3.20

<|outofrange|>

3.20

↵

3.20

↵↵

3.20

Activations Density 0.233%

the word "from."

No Comments

No Known Activations