INDEX

Explanations

mentions or discussions of the term "fake news" in text

oai_token-act-pair · gpt-3.5-turbo

references to "fake news."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

oples

-0.74

 Scher

-0.74

xus

-0.72

joining

-0.72

 kindred

-0.67

[|

-0.66

ental

-0.64

 Nile

-0.63

 skilled

-0.63

asse

-0.63

POSITIVE LOGITS

room

0.91

iness

0.84

rooms

0.84

feed

0.82

 headlines

0.78

caster

0.77

worthy

0.77

ãĤ±

0.76

icle

0.76

ashington

0.75

Activations Density 0.031%

mentions or discussions of the term "fake news" in text

references to "fake news."

No Comments

No Known Activations