INDEX

Explanations

references to "rules of thumb" or general guidelines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

erson

-0.16

ropolis

-0.14

ija

-0.14

iji

-0.14

ors

-0.14

obel

-0.14

idelberg

-0.14

lays

-0.14

 imageSize

-0.14

 Mess

-0.13

POSITIVE LOGITS

uxt

0.16

rule

0.16

onnen

0.15

 rule

0.15

æĶ»

0.14

yst

0.14

PIO

0.14

.nd

0.14

innen

0.14

_scalar

0.14

Activations Density 0.245%

references to "rules of thumb" or general guidelines

No Comments

No Known Activations

references to "rules of thumb" or general guidelines

No Comments

No Known Activations