INDEX

Explanations

phrases related to product ratings and recommendations

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

Gy

-0.68

MGM

-0.68

GH

-0.68

utt

-0.67

 groom

-0.66

lr

-0.66

kr

-0.65

 sled

-0.65

hub

-0.64

 marsh

-0.64

POSITIVE LOGITS

this

1.09

 this

0.98

This

0.88

 THIS

0.86

 This

0.84

these

0.76

THIS

0.75

ERAL

0.73

 largeDownload

0.71

 "$:/

0.71

Activations Density 0.041%

phrases related to product ratings and recommendations

No Comments

No Known Activations