INDEX

Explanations

references to family dynamics and interpersonal relationships

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

ivable

-0.66

utenberg

-0.65

 opponents

-0.65

IVERS

-0.65

 cumulative

-0.65

ATIONAL

-0.64

 convergence

-0.64

 proponents

-0.64

icipated

-0.63

uably

-0.61

POSITIVE LOGITS

 theirs

1.04

 whom

0.95

 kindly

0.88

 hers

0.87

 swoop

0.85

 whispering

0.82

 jealous

0.79

 upstairs

0.77

 babys

0.75

 beck

0.74

Activations Density 5.166%

references to family dynamics and interpersonal relationships

No Comments

No Known Activations