INDEX

Explanations

possessive forms and expressions of ownership

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

Ã«n

-0.21

ster

-0.17

’na

-0.17

’d

-0.15

ular

-0.15

liness

-0.15

’s

-0.15

ish

-0.15

sher

-0.15

em

-0.14

POSITIVE LOGITS

 Choice

0.22

aurus

0.21

-eye

0.20

sah

0.18

Eve

0.18

ÂĢÂĻ

0.17

chaft

0.17

'-

0.17

Choice

0.17

Activations Density 0.116%

possessive forms and expressions of ownership

No Comments

No Known Activations