INDEX

Explanations

data, results, status

np_max-act · gemini-2.0-flash

The neuron activates on JSON object key names (i.e. the field identifiers in JSON structures).

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

天天

-0.07

토

-0.07

 iterative

-0.07

 nuit

-0.07

_SH

-0.07

IRECTION

-0.06

 weakened

-0.06

舰

-0.06

FAST

-0.06

 adresse

-0.06

POSITIVE LOGITS

-search

0.07

-owner

0.06

отв

0.06

 دریافت

0.06

;")↵

0.06

 Skip

0.06

 servings

0.06

富

0.06

 Continue

0.06

Це

0.06

Activations Density 0.024%

data, results, status

The neuron activates on JSON object key names (i.e. the field identifiers in JSON structures).

No Comments

No Known Activations