INDEX

Explanations

keywords related to sporting events, particularly the term "shirt"

oai_token-act-pair · gpt-3.5-turbo

terms related to "redshirt" status in sports contexts

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

mercial

-0.75

 cannabin

-0.69

 IMAGES

-0.69

hran

-0.66

BIL

-0.65

atican

-0.64

sonian

-0.64

herical

-0.64

 VIDEOS

-0.63

perty

-0.63

POSITIVE LOGITS

oxide

0.74

otide

0.72

 Peaks

0.65

oche

0.65

opsis

0.63

issions

0.62

pine

0.61

oots

0.61

hee

0.60

xia

0.60

Activations Density 0.072%

keywords related to sporting events, particularly the term "shirt"

terms related to "redshirt" status in sports contexts

No Comments

No Known Activations