INDEX

Explanations

mentions of actions of blocking or being blocked (e.g., "Blocked", "Unblock", "Blocking")

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "blocked."

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

Sov

-0.72

shown

-0.67

gain

-0.67

 abund

-0.67

repre

-0.65

present

-0.64

rift

-0.64

 nucleus

-0.64

eah

-0.63

bons

-0.63

POSITIVE LOGITS

ãĤ´ãĥ³

0.82

ĵĺ

0.82

utsche

0.72

ogging

0.71

icer

0.68

wana

0.68

ESCO

0.67

iard

0.66

 Lists

0.65

jamin

0.64

Activations Density 0.022%

mentions of actions of blocking or being blocked (e.g., "Blocked", "Unblock", "Blocking")

occurrences of the word "blocked."

No Comments

No Known Activations