INDEX

Explanations

references to the term "Hyper" or related concepts in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oured

-0.81

ouring

-0.77

%]

-0.76

URES

-0.69

UES

-0.69

amental

-0.68

////

-0.68

çĶ°

-0.68

ivity

-0.67

POSITIVE LOGITS

visor

1.23

visors

1.16

bole

1.02

dimension

0.98

loop

0.79

LAN

0.75

closure

0.75

plex

0.74

rule

0.74

netic

0.73

Activations Density 0.004%

references to the term "Hyper" or related concepts in various contexts

No Comments

No Known Activations