INDEX

Explanations

classifications or designations of individuals and their professional titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ite

-0.17

itech

-0.16

ith

-0.15

als

-0.15

iform

-0.15

ines

-0.15

 prev

-0.15

abad

-0.15

ql

-0.14

ie

-0.14

POSITIVE LOGITS

’S

0.23

_EDEFAULT

0.19

'S

0.19

ING

0.18

/OR

0.17

phinx

0.16

ighth

0.15

AND

0.15

edir

0.15

ancel

0.15

Activations Density 0.173%

classifications or designations of individuals and their professional titles

No Comments

No Known Activations