INDEX

Explanations

mentions of the word "peak"

oai_token-act-pair · gpt-3.5-turbo

references to maximum levels or peaks 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 @ 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

icate

-0.75

oured

-0.74

ktop

-0.73

signed

-0.72

erie

-0.69

cheat

-0.69

onsense

-0.67

 parted

-0.66

 Cosponsors

-0.66

recated

-0.66

POSITIVE LOGITS

 erection

0.77

 peaks

0.73

urb

0.73

 heights

0.72

lite

0.71

 height

0.71

ceed

0.69

 occupancy

0.69

 capacity

0.69

imum

0.69

Activations Density 0.019%

mentions of the word "peak"

references to maximum levels or peaks in various contexts

No Comments

No Known Activations