INDEX

Explanations

references to reviewers and feedback-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

letic

-0.17

ema

-0.17

abad

-0.15

.Disabled

-0.15

pter

-0.15

enne

-0.14

iph

-0.14

raith

-0.14

uary

-0.13

 Phantom

-0.13

POSITIVE LOGITS

iqu

0.15

wend

0.14

repid

0.14

onas

0.14

 Coconut

0.14

obre

0.14

ewis

0.14

/compiler

0.14

dup

0.13

Activations Density 0.027%

references to reviewers and feedback-related terms

No Comments

No Known Activations