INDEX

Explanations

instances of the verb "have" and its variants indicating possession or experience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

à¸¶à¹Ī

-0.18

azar

-0.16

-0.15

 stÃ¡t

-0.15

ugen

-0.15

aat

-0.14

animated

-0.14

 rodi

-0.14

ARSE

-0.13

athers

-0.13

POSITIVE LOGITS

 such

0.21

nt

0.20

it

0.20

0.19

 finally

0.19

to

0.18

 this

0.18

 tons

0.18

no

0.18

 zero

0.18

Activations Density 0.310%

instances of the verb "have" and its variants indicating possession or experience

No Comments

No Known Activations