INDEX

Explanations

explanations or reasons within a context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "explain" in various contexts

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

illet

-0.81

luster

-0.76

ammy

-0.76

estial

-0.75

ches

-0.73

AUT

-0.68

ngth

-0.68

sembly

-0.67

nown

-0.67

ibaba

-0.67

POSITIVE LOGITS

why

1.12

WHY

1.06

ĸļ

0.93

why

0.91

cases

0.84

how

0.79

udic

0.77

 explan

0.75

 explanations

0.74

orial

0.74

Activations Density 0.035%

explanations or reasons within a context

instances of the word "explain" in various contexts

No Comments

No Known Activations