INDEX

Explanations

references to previous experiences or comparisons over time

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

icken

-0.79

bb

-0.69

fw

-0.69

bard

-0.69

okers

-0.69

motion

-0.68

okes

-0.68

externalActionCode

-0.68

tip

-0.68

mouth

-0.67

POSITIVE LOGITS

 previous

1.04

 usual

0.91

 norm

0.83

 baseline

0.80

 whence

0.80

 traditional

0.80

 typical

0.78

 scrimmage

0.77

 earlier

0.75

 conventional

0.74

Activations Density 0.042%

references to previous experiences or comparisons over time

No Comments

No Known Activations