INDEX

Explanations

words related to specific locations or structures, potentially related to a server or a specific place name

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "vern" in various contexts

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

 Beir

-0.65

ARC

-0.64

FTA

-0.64

CHR

-0.64

OHN

-0.62

ģĸ

-0.62

FT

-0.62

Wan

-0.58

Single

-0.58

 bout

-0.57

POSITIVE LOGITS

acular

1.27

ocity

0.99

issance

0.95

vern

0.94

sburg

0.92

ormal

0.90

okes

0.88

unci

0.86

ational

0.85

unciation

0.84

Activations Density 0.006%

words related to specific locations or structures, potentially related to a server or a specific place name

occurrences of the word "vern" in various contexts

No Comments

No Known Activations