INDEX

Explanations

Uncertainty and unknown information

np_max-act · gemini-2.0-flash

The neuron flags vague/hedging terms that signal uncertainty or non-specificity (e.g. “something,” “unclear,” “debate,” “perhaps,” etc.).

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

_strength

-0.07

routing

-0.06

.(*

-0.06

#{

-0.06

.:.:.:.:

-0.06

 troll

-0.06

Marco

-0.06

 cabbage

-0.06

Sou

-0.06

}"

-0.06

POSITIVE LOGITS

Email

0.07

 Gameplay

0.06

ện

0.06

Ordered

0.06

先

0.06

 NOTES

0.06

GetString

0.06

 getToken

0.06

yps

0.06

 соврем

0.06

Activations Density 0.028%

Uncertainty and unknown information

The neuron flags vague/hedging terms that signal uncertainty or non-specificity (e.g. “something,” “unclear,” “debate,” “perhaps,” etc.).

No Comments

No Known Activations

Uncertainty and unknown information

The neuron flags vague/hedging terms that signal uncertainty or non-specificity (e.g. “something,” “unclear,” “debate,” “perhaps,” etc.).

No Comments

No Known Activations