INDEX

Explanations

phrases related to online platforms or technological developments

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "co" and variations of "la" in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DERR

-0.65

 gems

-0.65

 Annotations

-0.64

 Hunger

-0.62

 employing

-0.61

REDACTED

-0.61

 Rampage

-0.61

 tones

-0.60

 Guilty

-0.60

 deprivation

-0.59

POSITIVE LOGITS

gging

1.14

gged

1.11

etooth

1.09

ggle

1.05

cks

1.05

opy

1.02

pper

1.01

ister

1.00

eps

0.98

pload

0.97

Activations Density 0.117%

phrases related to online platforms or technological developments

occurrences of the substring "co" and variations of "la" in the text

No Comments

No Known Activations