INDEX

Explanations

words containing the prefix "un-" followed by a word stem

oai_token-act-pair · gpt-3.5-turbo

negative prefixes related to lack or absence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Dynamics

-0.85

OPLE

-0.84

ãĥ¼ãĥĨãĤ£

-0.82

hetti

-0.79

 briefs

-0.75

anwhile

-0.74

jriwal

-0.69

Tut

-0.69

utical

-0.67

ORY

-0.67

POSITIVE LOGITS

balanced

1.22

assuming

1.16

confirmed

1.15

apolog

1.14

leased

1.12

availability

1.11

classified

1.11

cles

1.11

ifying

1.11

ruly

1.11

Activations Density 0.032%

words containing the prefix "un-" followed by a word stem

negative prefixes related to lack or absence

No Comments

No Known Activations