INDEX

Explanations

references to specific terms or concepts related to organizations or institutions

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

¸

-2.06

¿½

-1.76

¦

-1.74

Īĺ

-1.60

·

-1.60

 wrong

-1.52

bs

-1.51

inct

-1.48

¤

-1.47

¨

-1.45

POSITIVE LOGITS

oire

2.00

ophone

1.88

oria

1.72

ymphony

1.70

ori

1.69

oir

1.60

velopment

1.60

orship

1.59

urer

1.59

pace

1.58

Activations Density 0.097%

references to specific terms or concepts related to organizations or institutions

No Comments

No Known Activations