INDEX

Explanations

words related to specific geographical locations

oai_token-act-pair · gpt-3.5-turbo

references to geographical locations, specifically places related to "land."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 cumbers

-0.89

 tremend

-0.86

 millenn

-0.79

ollah

-0.68

 beware

-0.65

idium

-0.65

berus

-0.65

interest

-0.64

 punishing

-0.64

 undermin

-0.63

POSITIVE LOGITS

scape

1.12

marks

0.88

ward

0.86

gra

0.84

estine

0.84

 Park

0.81

mark

0.79

raft

0.78

 Acquisition

0.77

grave

0.77

Activations Density 0.017%

words related to specific geographical locations

references to geographical locations, specifically places related to "land."

No Comments

No Known Activations