INDEX

Explanations

occurrences of specific names or identifiers

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ascript

-0.90

ividual

-0.85

isite

-0.80

raltar

-0.76

ilts

-0.76

erest

-0.75

irlf

-0.75

gerald

-0.73

lections

-0.72

berus

-0.70

POSITIVE LOGITS

Pai

0.93

Dur

0.76

 Kumar

0.74

à¦

0.72

ullah

0.72

Rai

0.70

ji

0.70

 Chou

0.70

 Bang

0.69

 Gandhi

0.68

Activations Density 0.018%

occurrences of specific names or identifiers

No Comments

No Known Activations