INDEX

Explanations

names of sports teams and players

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

nces

-0.81

tml

-0.76

utic

-0.72

 Ples

-0.72

judicial

-0.67

uters

-0.66

ancies

-0.65

schild

-0.65

milo

-0.64

iaries

-0.64

POSITIVE LOGITS

 Athletics

1.04

 Stadium

0.96

 Fans

0.94

 fans

0.93

 rookie

0.92

FC

0.88

 Athletic

0.88

 fanbase

0.86

 Offense

0.86

 teammate

0.85

Activations Density 2.493%

names of sports teams and players

No Comments

No Known Activations