INDEX

Explanations

phrases or expressions indicating confusion or absurdity

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

rique

-0.70

ahime

-0.69

rica

-0.68

elist

-0.68

emale

-0.68

ngth

-0.66

rican

-0.65

lean

-0.65

essor

-0.64

 Citation

-0.64

POSITIVE LOGITS

 except

0.86

together

0.84

 toget

0.82

 moot

0.78

 together

0.73

 happening

0.70

 transpired

0.66

 revolves

0.65

 undone

0.64

usional

0.64

Activations Density 0.094%

phrases or expressions indicating confusion or absurdity

No Comments

No Known Activations