INDEX

Explanations

error correction

np_max-act · gemini-2.0-flash

The neuron activates on mentions of reinforcement‐learning concepts (e.g. “reinforcement learning,” “RLHF,” rewards/penalties).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Episode

-0.06

 rubber

-0.06

_shop

-0.06

Lex

-0.06

 Bulld

-0.06

 sales

-0.06

Words

-0.06

developer

-0.06

 mountains

-0.06

SetText

-0.06

POSITIVE LOGITS

Median

0.07

.wp

0.07

cds

0.07

適用

0.06

'['

0.06

_IMG

0.06

 walked

0.06

 kell

0.06

	RTLU

0.06

taient

0.06

Activations Density 0.021%

error correction

The neuron activates on mentions of reinforcement‐learning concepts (e.g. “reinforcement learning,” “RLHF,” rewards/penalties).

No Comments

No Known Activations