INDEX

Explanations

mentions or references to TV channels

oai_token-act-pair · gpt-3.5-turbo

references to television network names and their associated programs or news reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

leans

-0.80

hammad

-0.77

hot

-0.72

uten

-0.71

hend

-0.68

cific

-0.67

ometric

-0.66

-0.65

currency

-0.64

 otherwise

-0.63

POSITIVE LOGITS

ILLE

1.42

 Everywhere

0.89

NZ

0.88

Sav

0.83

trop

0.82

OS

0.82

 Broadcasting

0.81

 Channel

0.79

LV

0.78

 Shows

0.76

Activations Density 0.022%

mentions or references to TV channels

references to television network names and their associated programs or news reports

No Comments

No Known Activations