INDEX

Explanations

phrases related to political discussions and policies

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

irez

-0.68

 Presence

-0.65

 Prix

-0.64

 redesign

-0.63

umption

-0.61

racuse

-0.60

ulas

-0.60

ipation

-0.58

pedia

-0.58

 irresist

-0.58

POSITIVE LOGITS

 whatsoever

0.89

THING

0.72

worthiness

0.70

 sudden

0.70

us

0.68

those

0.66

affles

0.65

course

0.65

these

0.64

llor

0.64

Activations Density 0.043%

phrases related to political discussions and policies

No Comments

No Known Activations