INDEX

Explanations

references to additional or supplementary concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 reconciliation

-0.15

angered

-0.14

prung

-0.14

 runaway

-0.14

ESIS

-0.14

 cÃłng

-0.14

rypton

-0.14

upy

-0.14

 Dare

-0.13

ROP

-0.13

POSITIVE LOGITS

quier

0.16

gether

0.15

thing

0.15

things

0.14

avel

0.14

THREAD

0.14

Mor

0.14

dut

0.14

antity

0.14

 things

0.14

Activations Density 0.010%

references to additional or supplementary concepts

No Comments

No Known Activations