INDEX

Explanations

mentions of ranches

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lihood

-0.77

 dism

-0.65

ABLE

-0.61

 chromos

-0.60

 Crunch

-0.59

 Crimson

-0.58

 Eucl

-0.57

 Wong

-0.56

 Wonder

-0.55

 Grail

-0.55

POSITIVE LOGITS

 ranch

0.96

 Bundy

0.92

yard

0.90

lands

0.87

yards

0.86

ns

0.80

ÃŃa

0.77

ing

0.77

esian

0.76

arious

0.76

Activations Density 0.028%

mentions of ranches

No Comments

No Known Activations