INDEX

Explanations

words related to mental processes and activities involving analysis and organization

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

airs

-0.74

draw

-0.72

guard

-0.71

den

-0.71

VID

-0.68

spot

-0.67

REL

-0.66

croft

-0.65

POST

-0.65

occup

-0.64

POSITIVE LOGITS

imore

0.87

ulate

0.83

 oneself

0.83

 into

0.80

escent

0.76

ify

0.71

 INTO

0.70

imize

0.70

 everything

0.69

 entire

0.68

Activations Density 1.551%

words related to mental processes and activities involving analysis and organization

No Comments

No Known Activations