INDEX

Explanations

the word "explain" and possibly words related to scientific explanations or theories.

oai_token-act-pair · gemini-1.5-flash Triggered by @johnny

explanations or theories

oai_token-act-pair · gpt-3.5-turbo

phrases related to providing explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

engeance

-0.83

nown

-0.79

illet

-0.77

inal

-0.76

jet

-0.76

thritis

-0.74

nir

-0.72

naissance

-0.72

ascus

-0.71

atri

-0.70

POSITIVE LOGITS

why

1.15

WHY

1.10

 explanations

0.98

ĸļ

0.93

 explan

0.93

 explain

0.93

 Explain

0.92

why

0.92

 explains

0.87

urated

0.86

Activations Density 0.020%

the word "explain" and possibly words related to scientific explanations or theories.

explanations or theories

phrases related to providing explanations

No Comments

No Known Activations