INDEX

Explanations

statements asserting factual information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

thus

-0.15

ensis

-0.15

jac

-0.14

nek

-0.14

foundland

-0.14

ould

-0.14

ULD

-0.13

Î±Î¹Î½

-0.13

/package

-0.13

acht

-0.13

POSITIVE LOGITS

 fact

0.23

itious

0.23

uality

0.18

ually

0.18

0.15

ease

0.14

fact

0.14

Ã©o

0.14

TA

0.14

arding

0.13

Activations Density 0.034%

statements asserting factual information

No Comments

No Known Activations