INDEX

Explanations

terms related to awards, events, and acknowledgments in a competitive context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

onto

-0.56

 speedy

-0.56

whatever

-0.55

 sinful

-0.55

 opposition

-0.55

criminal

-0.54

entimes

-0.54

 hers

-0.54

atever

-0.54

luster

-0.53

POSITIVE LOGITS

*:

0.89

*)

0.89

 Â·

0.87

+++

0.78

ãĢĳ

0.77

 refers

0.74

::::::::

0.74

âĦ¢:

0.73

 typo

0.73

-->

0.72

Activations Density 0.292%

terms related to awards, events, and acknowledgments in a competitive context

No Comments

No Known Activations