INDEX

Explanations

tags indicating categories or labels within the text

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

ĳ

-3.96

ĸ´

-3.60

ĥ½

-3.56

ħ

-3.51

-3.47

↵↵

-3.47

<|outofrange|>

-3.47

↵

-3.47

POSITIVE LOGITS

gered

1.75

read

1.72

zilla

1.72

lia

1.71

lane

1.70

liament

1.68

gart

1.67

gun

1.64

alin

1.58

ied

1.54

Activations Density 0.009%

tags indicating categories or labels within the text

No Comments

No Known Activations