INDEX

Explanations

general references to the concept of "all" or totality

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

thren

-0.67

rients

-0.67

ritic

-0.65

reat

-0.63

chieve

-0.63

elve

-0.62

atro

-0.61

pperc

-0.61

eele

-0.61

atches

-0.60

POSITIVE LOGITS

uding

1.04

usion

0.99

uring

0.96

ocating

0.96

ocated

0.82

 except

0.79

usions

0.78

ayed

0.77

 encomp

0.76

udes

0.75

Activations Density 0.019%

general references to the concept of "all" or totality

No Comments

No Known Activations