INDEX

Explanations

terms related to interviews

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "interview"

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

axy

-0.87

¶ħ

-0.82

math

-0.79

mil

-0.74

cil

-0.74

hover

-0.70

ciplinary

-0.69

fal

-0.68

¬¼

-0.68

sels

-0.68

POSITIVE LOGITS

ees

1.16

 interviews

1.07

 interview

1.04

 Interview

0.96

Interview

0.90

ioned

0.87

 transcripts

0.85

ee

0.84

 Transcript

0.83

 interviewing

0.82

Activations Density 0.028%

terms related to interviews

occurrences of the word "interview"

No Comments

No Known Activations