INDEX

Explanations

terms related to predictions and forecasting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 zuge

-0.34

eventType

-0.33

setDuration

-0.32

 Dowling

-0.32

Composição

-0.32

atmosphère

-0.32

Ằ

-0.31

 Montes

-0.31

ående

-0.31

POSITIVE LOGITS

InputTagHelper

0.72

 prediction

0.71

Prediction

0.69

 виправивши

0.68

prediction

0.66

 predictions

0.66

 Prediction

0.64

 doInBackground

0.62

IBOutlet

0.62

IndentedString

0.61

Activations Density 0.067%

terms related to predictions and forecasting

No Comments

No Known Activations