INDEX

Explanations

references to programming elements and constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 forth

-0.15

_consts

-0.15

holm

-0.14

ł

-0.14

segue

-0.14

stown

-0.14

assin

-0.14

å¾ģ

-0.13

chio

-0.13

rais

-0.13

POSITIVE LOGITS

anner

0.15

eyJ

0.14

eger

0.14

 Chains

0.14

ruba

0.14

 Berk

0.13

dives

0.13

 ComVisible

0.13

enn

0.13

ilar

0.13

Activations Density 0.536%

references to programming elements and constructs

No Comments

No Known Activations