INDEX

Explanations

numbers followed by a value of 'or' in a context of comparison or range

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

ONSORED

-0.70

WER

-0.62

RPG

-0.62

 Candidate

-0.60

RP

-0.60

CRE

-0.58

MODE

-0.57

INS

-0.57

CLA

-0.56

REP

-0.56

POSITIVE LOGITS

 thirds

0.85

teenth

0.85

undreds

0.81

thirds

0.81

dozen

0.80

elfth

0.80

quarters

0.80

liest

0.79

tons

0.78

seven

0.78

Activations Density 1.094%

numbers followed by a value of 'or' in a context of comparison or range

No Comments

No Known Activations