INDEX

Explanations

references to decisions, actions, or opinions expressed with a negative emotional tone, possibly related to criticism or rejection

oai_token-act-pair · gpt-3.5-turbo

comparisons or contrasts between past and present conditions or roles

oai_token-act-pair · gpt-3.5-turbo Triggered by @danbraun

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

verty

-0.68

Led

-0.67

olate

-0.64

aez

-0.62

atures

-0.60

enos

-0.59

soType

-0.58

uben

-0.57

uid

-0.57

ixtures

-0.57

POSITIVE LOGITS

 altogether

1.21

 entirely

0.94

 outright

0.80

 attempts

0.73

 precon

0.69

 nonsense

0.68

 pret

0.67

 objection

0.67

 objections

0.67

 attempt

0.67

Activations Density 14.737%

references to decisions, actions, or opinions expressed with a negative emotional tone, possibly related to criticism or rejection

comparisons or contrasts between past and present conditions or roles

No Comments

No Known Activations