INDEX

Explanations

terms related to emotional states and conditions, particularly focusing on descriptors of feelings and circumstances

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

too

-0.84

Too

-0.79

Too

-0.74

too

-0.72

TOO

-0.68

 terlalu

-0.66

 слишком

-0.57

TOO

-0.56

troppo

-0.55

 demasiado

-0.54

POSITIVE LOGITS

حوالہ

0.42

ocino

0.41

 chi̍t

0.41

weta

0.39

Портал

0.39

født

0.39

uries

0.38

Попис

0.38

 vetro

0.37

 Riproduzione

0.37

Activations Density 0.046%

terms related to emotional states and conditions, particularly focusing on descriptors of feelings and circumstances

No Comments

No Known Activations