INDEX

Explanations

words ending with the letters 'ts' and 'es'

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

ously

-0.69

 decom

-0.59

 Netanyahu

-0.59

 visitation

-0.58

 evolution

-0.57

 manipulation

-0.57

BDS

-0.56

 resistance

-0.55

 nerv

-0.55

 Modi

-0.55

POSITIVE LOGITS

hire

1.26

poon

1.20

ilver

1.12

pring

1.12

omething

1.12

dale

1.12

bury

1.08

poons

1.04

cript

1.02

hare

1.02

Activations Density 0.035%

words ending with the letters 'ts' and 'es'

No Comments

No Known Activations