INDEX

Explanations

instances of the word "to" used in various contexts

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

ãĤ¦ãĤ¹

-0.83

processing

-0.81

Reviewed

-0.79

chlor

-0.71

urized

-0.69

sett

-0.68

lime

-0.68

imon

-0.66

intensive

-0.64

 Balanced

-0.64

POSITIVE LOGITS

 coincide

1.35

 resemble

1.13

 introduce

1.07

 recognise

1.05

be

1.01

 meet

0.97

 belong

0.96

 mention

0.96

 become

0.96

 reside

0.95

Activations Density 0.017%

instances of the word "to" used in various contexts

No Comments

No Known Activations