INDEX

Explanations

words related to answering questions

oai_token-act-pair · gpt-3.5-turbo

phrases that involve responding to questions

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

zinski

-0.76

chin

-0.76

 Nanto

-0.74

heric

-0.70

erker

-0.69

 Vengeance

-0.68

ufact

-0.68

ovie

-0.67

robat

-0.65

akin

-0.65

POSITIVE LOGITS

ysis

1.13

answer

0.98

 answ

0.98

 answering

0.90

swers

0.88

 answered

0.84

 questions

0.82

Answer

0.81

 Questions

0.79

ĵĺ

0.77

Activations Density 0.024%

words related to answering questions

phrases that involve responding to questions

No Comments

No Known Activations