INDEX

Explanations

references to specific medical conditions and their related keywords

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

vic

-0.73

 Canaver

-0.72

å°Ĩ

-0.62

 prost

-0.59

 STATS

-0.59

 horm

-0.58

 imbalance

-0.58

 BASE

-0.57

Prev

-0.56

 modulation

-0.54

POSITIVE LOGITS

culosis

0.92

zee

0.77

ongo

0.72

ville

0.69

cular

0.69

asa

0.69

zees

0.66

 Mongo

0.66

hower

0.65

itus

0.65

Activations Density 0.021%

references to specific medical conditions and their related keywords

No Comments

No Known Activations