INDEX

Explanations

phrases related to opposing views or choices

oai_token-act-pair · gpt-3.5-turbo

concepts related to divergence and polarity in viewpoints

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.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

 Annotations

-0.73

Delivery

-0.71

hey

-0.70

 Regist

-0.64

Register

-0.64

Tickets

-0.63

urga

-0.62

Install

-0.62

ogo

-0.61

Emb

-0.60

POSITIVE LOGITS

 oppos

1.08

 sexes

1.07

 genders

0.95

 viewpoints

0.91

 extremes

0.90

inational

0.86

 twins

0.84

 halves

0.84

 disparate

0.84

 perspectives

0.84

Activations Density 0.074%

phrases related to opposing views or choices

concepts related to divergence and polarity in viewpoints

No Comments

No Known Activations