INDEX

Explanations

terms related to the company "Google"

oai_token-act-pair · gpt-3.5-turbo

mentions of the company "Google."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

alid

-0.79

mble

-0.77

ROR

-0.77

ivari

-0.69

Wem

-0.68

rone

-0.67

OHN

-0.66

stood

-0.66

quez

-0.65

swing

-0.65

POSITIVE LOGITS

plex

0.88

 Scholar

0.88

geist

0.87

 Doodle

0.82

 Google

0.80

 searches

0.76

 Voice

0.75

Google

0.75

 LinkedIn

0.74

 Lunar

0.74

Activations Density 0.012%

terms related to the company "Google"

mentions of the company "Google."

No Comments

No Known Activations