INDEX

Explanations

statements or text introductions that begin with the word "Because"

oai_token-act-pair · gpt-3.5-turbo

causal relationships or explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lem

-0.79

ymph

-0.75

yan

-0.70

intern

-0.68

Gas

-0.66

scr

-0.63

age

-0.63

åĤ

-0.61

shr

-0.61

MC

-0.61

POSITIVE LOGITS

rely

1.04

akening

0.72

urers

0.69

pite

0.69

uesday

0.68

pread

0.64

ecause

0.63

ertodd

0.63

 hindsight

0.63

awaru

0.62

Activations Density 0.054%

statements or text introductions that begin with the word "Because"

causal relationships or explanations

No Comments

No Known Activations