INDEX

Explanations

sections of code or comments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ActionCreators

-0.19

-0.16

_nullable

-0.15

ERA

-0.15

ç´

-0.14

kest

-0.14

naires

-0.14

Å¡il

-0.14

pend

-0.14

ware

-0.13

POSITIVE LOGITS

0.15

assin

0.15

eding

0.15

Bij

0.14

ALSE

0.14

lg

0.13

mix

0.13

ounters

0.13

Activations Density 0.044%

sections of code or comments

No Comments

No Known Activations