INDEX

Explanations

attends to citations marked with "ibid" from low-related tokens

oai_attention-head · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_0/width_16k/average_l0_104

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.attn.hook_z

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

Head Attr Weights

0:0.03

1:0.04

2:0.04

3:0.32

4:0.10

5:0.04

6:0.30

7:0.09

Negative Logits

Дереккөздер

-0.59

 myſelf

-0.54

 itſelf

-0.53

 purpoſe

-0.50

ſelf

-0.47

 houſe

-0.46

 Majefty

-0.46

ſelves

-0.46

 pleaſure

-0.46

 faſt

-0.45

POSITIVE LOGITS

providedIn

0.33

setCancelable

0.29

 JsonObject

0.27

ú

0.26

雅

0.26

TestingModule

0.26

ort

0.26

 оче

0.26

mor

0.25

 proprie

0.25

Activations Density 0.039%

attends to citations marked with "ibid" from low-related tokens

No Comments

No Known Activations