INDEX

Explanations

phrases related to data processing, latency, and performance

oai_token-act-pair · gpt-3.5-turbo

technical terms and concepts related to performance optimizations and computing efficiency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.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

 Ally

-0.86

waters

-0.78

brand

-0.76

 Holden

-0.73

arag

-0.72

approved

-0.71

Male

-0.70

 Waters

-0.69

worn

-0.69

ogie

-0.69

POSITIVE LOGITS

 throughput

1.69

 latency

1.62

CPU

1.54

cpu

1.53

 caching

1.49

 computation

1.45

 workload

1.42

 bottleneck

1.42

 compute

1.41

 runtime

1.41

Activations Density 0.582%

phrases related to data processing, latency, and performance

technical terms and concepts related to performance optimizations and computing efficiency

No Comments

No Known Activations