INDEX

Explanations

the word "reasons" and related phrases indicating justifications or motivations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yss

-0.76

ibaba

-0.72

boro

-0.70

oba

-0.69

externalActionCode

-0.63

esc

-0.61

annis

-0.61

enged

-0.61

opy

-0.60

eng

-0.60

POSITIVE LOGITS

abl

0.99

 reasons

0.99

 unrelated

0.87

 alone

0.82

 Reasons

0.78

why

0.78

 inexplicable

0.77

atical

0.77

 unexplained

0.76

 unspecified

0.74

Activations Density 0.022%

the word "reasons" and related phrases indicating justifications or motivations

No Comments

No Known Activations