INDEX

Explanations

references to the television show "Saturday Night Live"

oai_token-act-pair · gpt-3.5-turbo

references to "Saturday Night Live" and notable television shows

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

staking

-0.72

andr

-0.70

uckland

-0.69

rastructure

-0.66

oiler

-0.65

wid

-0.65

 warr

-0.65

skilled

-0.64

ccording

-0.61

decl

-0.61

POSITIVE LOGITS

 Presents

0.95

 Channel

0.92

 aired

0.86

Radio

0.83

Â®

0.83

 Episode

0.83

 fame

0.81

 airs

0.78

 Decoder

0.78

 television

0.75

Activations Density 0.077%

references to the television show "Saturday Night Live"

references to "Saturday Night Live" and notable television shows

No Comments

No Known Activations