INDEX

Explanations

numerical values, particularly in a context related to measurement or evaluation

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

 Skydragon

-0.80

uyomi

-0.77

 conduc

-0.76

 livest

-0.74

 juggling

-0.73

 myster

-0.68

 headlights

-0.68

 puzz

-0.68

hner

-0.68

 contrace

-0.67

POSITIVE LOGITS

rican

0.98

Ñĭ

0.95

ERN

0.91

ric

0.90

Ø³

0.89

rik

0.89

ÙĪ

0.89

rique

0.87

Ñĥ

0.86

ãĥ¼

0.85

Activations Density 0.003%

numerical values, particularly in a context related to measurement or evaluation

No Comments

No Known Activations