INDEX

Explanations

phrases that indicate new beginnings or events starting

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

inately

-0.89

ategor

-0.75

 downed

-0.66

pedia

-0.63

 outgoing

-0.61

 Quarterly

-0.60

pell

-0.59

 persist

-0.58

verbs

-0.58

peer

-0.58

POSITIVE LOGITS

nings

1.05

liest

1.01

 anew

0.95

ablishment

0.76

 thereof

0.75

 beginnings

0.71

uring

0.64

 phase

0.64

 realization

0.63

date

0.63

Activations Density 0.027%

phrases that indicate new beginnings or events starting

No Comments

No Known Activations