INDEX

Explanations

phrases related to disregarding or ignoring various things

oai_token-act-pair · gpt-3.5-turbo

instances of ignoring or overlooking significant issues or details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

reau

-0.87

artney

-0.82

raq

-0.74

ificent

-0.72

olate

-0.68

ONSORED

-0.68

arta

-0.67

soDeliveryDate

-0.67

 roundup

-0.66

nai

-0.65

POSITIVE LOGITS

 altogether

1.11

 entirely

0.85

 pesky

0.79

 nuance

0.77

 mantle

0.77

 distinctions

0.77

 outright

0.76

 distinction

0.74

 nuances

0.74

 fact

0.74

Activations Density 0.367%

phrases related to disregarding or ignoring various things

instances of ignoring or overlooking significant issues or details

No Comments

No Known Activations