INDEX

Explanations

phrases related to influence and impact

oai_token-act-pair · gpt-3.5-turbo 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Cosponsors

-0.80

ML

-0.72

GW

-0.71

 Strikes

-0.64

ALLY

-0.63

aneous

-0.63

NESS

-0.62

ML

-0.62

GW

-0.61

uits

-0.61

POSITIVE LOGITS

 undone

1.10

 untouched

1.07

 footprints

1.06

 intact

1.04

 unanswered

0.99

 residue

0.96

 behind

0.95

 unfinished

0.93

 voic

0.92

 unexpl

0.91

Activations Density 0.134%

phrases related to influence and impact

No Comments

No Known Activations