INDEX

Explanations

phrases related to tuning into something, such as TV shows, events, or specific channels

oai_token-act-pair · gpt-3.5-turbo

instances of the word "tune" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

perse

-0.72

anmar

-0.68

inals

-0.67

lain

-0.66

pha

-0.65

ylum

-0.64

Lak

-0.63

soever

-0.62

ãĤ¨ãĥ«

-0.60

Magikarp

-0.59

POSITIVE LOGITS

 tuned

0.93

tun

0.92

 tune

0.90

 tunes

0.86

 horns

0.86

 tuning

0.84

 Tune

0.84

tun

0.80

eness

0.78

Tun

0.76

Activations Density 0.056%

phrases related to tuning into something, such as TV shows, events, or specific channels

instances of the word "tune" and its variations

No Comments

No Known Activations