INDEX

Explanations

attends to "seem" and similar verbs from tokens that express judgment or perception about those observations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.attn.hook_z

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.10

1:0.11

2:0.14

3:0.06

4:0.05

5:0.01

6:0.08

7:0.41

Negative Logits

 dAtA

-0.40

 Normdatei

-0.36



-0.35

iprot

-0.34

 kasarigan

-0.32

ArgsConstructor

-0.32

 дописавши

-0.31

adaptiveStyles

-0.31

BufferException

-0.30

Geplaatst

-0.30

POSITIVE LOGITS

__;

0.27

 Felder

0.25

welche

0.24

OVID

0.24

 heller

0.23

acrylamide

0.23

هي

0.23

 then

0.23

 Sorgen

0.22

 Profile

0.22

Activations Density 0.275%

attends to "seem" and similar verbs from tokens that express judgment or perception about those observations

No Comments

No Known Activations