INDEX

Explanations

occurrences of the word "in" and phrases related to location or context

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

awatts

-0.73

MpServer

-0.72

ById

-0.69

WER

-0.65

asher

-0.65

 ASAP

-0.64

ãĥĺãĥ©

-0.61

çļ

-0.60

 descendants

-0.60

lly

-0.60

POSITIVE LOGITS

 captivity

0.95

 researching

0.86

 vacation

0.81

 touring

0.80

 Tours

0.77

boarding

0.74

 tour

0.73

 filming

0.70

])

0.68

 exile

0.67

Activations Density 0.039%

occurrences of the word "in" and phrases related to location or context

No Comments

No Known Activations