INDEX

Explanations

phrases related to human interactions and social dynamics

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

pora

-0.93

only

-0.80

inducing

-0.74

successfully

-0.71

impl

-0.71

rimination

-0.68

egal

-0.67

etheus

-0.67

agically

-0.67

tie

-0.67

POSITIVE LOGITS

 coincidence

0.89

 luck

0.80

 plain

0.76

 curiosity

0.72

 darn

0.71

 cents

0.70

ifiable

0.69

 warmed

0.68

 PowerPoint

0.68

 Facts

0.67

Activations Density 0.866%

phrases related to human interactions and social dynamics

No Comments

No Known Activations