INDEX

Explanations

phrases related to seeking evidence or proof of causation

oai_token-act-pair · gpt-3.5-turbo

concepts and discussions surrounding causation and evidence

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

undai

-0.75

mount

-0.74

eatures

-0.73

awar

-0.71

yrinth

-0.69

mounted

-0.68

hap

-0.67

artney

-0.67

oglu

-0.65

ibaba

-0.64

POSITIVE LOGITS

 anymore

0.87

ãĢĤ

0.85

 uttered

0.85

 thereof

0.79

 whatsoever

0.75

worthiness

0.75

 correctness

0.75

 truths

0.72

ãģ£

0.71

.</

0.69

Activations Density 0.477%

phrases related to seeking evidence or proof of causation

concepts and discussions surrounding causation and evidence

No Comments

No Known Activations