INDEX

Explanations

numerical options or choices 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_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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aths

-0.76

mouth

-0.76

mination

-0.73

culosis

-0.72

thing

-0.71

ended

-0.69

rowth

-0.68

DEBUG

-0.68

kamp

-0.67

ibur

-0.66

POSITIVE LOGITS

 amongst

0.94

 assorted

0.92

 among

0.91

 list

0.88

 randomly

0.88

 various

0.83

 lists

0.82

 alphabet

0.81

 random

0.80

 repertoire

0.79

Activations Density 0.107%

numerical options or choices in various contexts

No Comments

No Known Activations