INDEX

Explanations

terms related to substitution or alternatives

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

raq

-0.95

jah

-0.85

-0.79

UGE

-0.77

eta

-0.76

apa

-0.75

vern

-0.73

ucket

-0.72

achine

-0.72

iland

-0.71

POSITIVE LOGITS

 limbs

0.77

 therapies

0.76

mentation

0.75

 therapy

0.75

 replacement

0.71

ding

0.70

 replacements

0.69

 parts

0.69

aneous

0.69

 casing

0.68

Activations Density 0.008%

terms related to substitution or alternatives

No Comments

No Known Activations