INDEX

Explanations

occurrences of the letter "E"

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

 Anthem

-0.73

 Bleach

-0.69

wagen

-0.69

 Kenobi

-0.66

rans

-0.64

 internationally

-0.61

 illusion

-0.59

 juggling

-0.59

 Nirvana

-0.58

raints

-0.58

POSITIVE LOGITS

nerg

1.15

ASY

1.14

fficient

1.12

ighty

1.12

tymology

1.08

isner

1.08

cosystem

1.07

lements

1.06

TERN

1.05

rect

1.05

Activations Density 0.023%

occurrences of the letter "E"

No Comments

No Known Activations